Audio¶
client.audio groups speech synthesis (TTS), recognition (ASR), realtime
WebSocket sessions and Qwen Live, plus voice/vocabulary customization.
Audio surface¶
Audio
¶
Audio(client)
Sync audio surface.
Source code in dashscope/resources/audio/__init__.py
175 176 177 178 179 180 181 182 183 | |
AsyncAudio
¶
AsyncAudio(client)
Async audio surface.
Only batch transcription is exposed here; the ws realtime resources
(asr.recognize, tts, realtime) are sync-only by design
(design doc M3 3.5).
Source code in dashscope/resources/audio/__init__.py
194 195 196 197 198 199 200 | |
Speech synthesis (TTS)¶
Tts
¶
Tts(client)
Speech synthesis (one-shot + streaming) over websocket.
Source code in dashscope/resources/audio/tts.py
291 292 293 | |
connect
¶
connect(
*,
model: str = "cosyvoice-v1",
voice: str = "longxiaochun",
format: str = "mp3",
sample_rate: int = 22050,
volume: Optional[int] = None,
rate: Optional[float] = None,
pitch: Optional[float] = None,
language_hints: Optional[List[str]] = None,
enable_ssml: Optional[bool] = None,
parameters: Optional[Dict[str, Any]] = None,
ws_url: Optional[str] = None,
connector: Optional[Connector] = None,
connection: Optional[WsConnection] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None
) -> SynthesisSession
Create a duplex session; enter its context or call start().
With connection= the session runs as one task on the shared
socket and claims its binary downlink (one active synthesis per
connection).
Source code in dashscope/resources/audio/tts.py
295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 | |
synthesize
¶
synthesize(
text: str,
*,
model: str = "cosyvoice-v1",
voice: str = "longxiaochun",
format: str = "mp3",
sample_rate: int = 22050,
volume: Optional[int] = None,
rate: Optional[float] = None,
pitch: Optional[float] = None,
language_hints: Optional[List[str]] = None,
enable_ssml: Optional[bool] = None,
parameters: Optional[Dict[str, Any]] = None,
ws_url: Optional[str] = None,
connector: Optional[Connector] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None
) -> bytes
Synthesize the whole text; returns the concatenated audio.
Source code in dashscope/resources/audio/tts.py
343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 | |
stream
¶
stream(
text: str,
*,
model: str = "cosyvoice-v1",
voice: str = "longxiaochun",
format: str = "mp3",
sample_rate: int = 22050,
volume: Optional[int] = None,
rate: Optional[float] = None,
pitch: Optional[float] = None,
language_hints: Optional[List[str]] = None,
enable_ssml: Optional[bool] = None,
parameters: Optional[Dict[str, Any]] = None,
ws_url: Optional[str] = None,
connector: Optional[Connector] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None
) -> Iterator[bytes]
Yield audio chunks as they arrive.
Source code in dashscope/resources/audio/tts.py
387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 | |
SynthesisSession
¶
SynthesisSession(
client,
*,
model: str = "cosyvoice-v1",
voice: str = "longxiaochun",
format: str = "mp3",
sample_rate: int = 22050,
volume: Optional[int] = None,
rate: Optional[float] = None,
pitch: Optional[float] = None,
language_hints: Optional[List[str]] = None,
enable_ssml: Optional[bool] = None,
parameters: Optional[Dict[str, Any]] = None,
ws_url: Optional[str] = None,
connector: Optional[Connector] = None,
connection: Optional[WsConnection] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None
)
Duplex synthesis with concurrent text input and one result iterator.
Source code in dashscope/resources/audio/tts.py
88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 | |
session
property
¶
session: Union[WsSession, WsTask]
Underlying websocket session, including its task_id.
start
¶
start() -> 'SynthesisSession'
Open the connection and wait for the task-started handshake.
Source code in dashscope/resources/audio/tts.py
152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 | |
send_text
¶
send_text(text: str) -> None
Send another text fragment; not allowed after stop or cancel.
Source code in dashscope/resources/audio/tts.py
189 190 191 192 193 194 195 196 197 198 199 200 | |
flush
¶
flush() -> None
Request synthesis of buffered text without ending input.
Source code in dashscope/resources/audio/tts.py
202 203 204 205 206 207 208 | |
stop
¶
stop() -> None
End input without blocking for results; keep reading afterward.
Source code in dashscope/resources/audio/tts.py
210 211 212 213 214 215 216 | |
cancel
¶
cancel() -> None
Cancel once, even after stop, before a terminal event arrives.
Source code in dashscope/resources/audio/tts.py
218 219 220 221 222 223 224 | |
events
¶
events() -> (
Generator[Union[Dict[str, Any], bytes], None, None]
)
Yield JSON and audio through task-finished, raising on failure.
Source code in dashscope/resources/audio/tts.py
233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 | |
audio_chunks
¶
audio_chunks() -> Iterator[bytes]
Yield only audio, retaining JSON metadata on this session.
Source code in dashscope/resources/audio/tts.py
266 267 268 269 270 271 272 273 274 | |
close
¶
close() -> None
Close the connection; a closed session cannot be restarted.
With a shared WsConnection this only detaches the task; the
socket belongs to the connection and stays open.
Source code in dashscope/resources/audio/tts.py
276 277 278 279 280 281 282 283 284 285 | |
HttpTts
¶
HttpTts(client)
client.audio.tts.http -- HTTP synthesis for both families.
Source code in dashscope/resources/audio/http_tts.py
642 643 | |
synthesize
¶
synthesize(
text: str,
*,
model: str,
voice: str,
format: Optional[str] = None,
sample_rate: Optional[int] = None,
language_type: Optional[str] = None,
parameters: Optional[Dict[str, Any]] = None,
extra_input: Optional[Dict[str, Any]] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[False] = False
) -> HttpSynthesis
synthesize(
text: str,
*,
model: str,
voice: str,
format: Optional[str] = None,
sample_rate: Optional[int] = None,
language_type: Optional[str] = None,
parameters: Optional[Dict[str, Any]] = None,
extra_input: Optional[Dict[str, Any]] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[True]
) -> "APIResponse"
synthesize(
text: str,
*,
model: str,
voice: str,
format: Optional[str] = None,
sample_rate: Optional[int] = None,
language_type: Optional[str] = None,
parameters: Optional[Dict[str, Any]] = None,
extra_input: Optional[Dict[str, Any]] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: bool = False
) -> Union[HttpSynthesis, "APIResponse"]
Synthesize in one POST; the audio comes back as a URL.
raw=True returns the :class:~dashscope.transport.APIResponse
before any error check, so a caller can read the status, headers and
elapsed of the response or re-parse it itself. The non-raw path
keeps response_body's envelope: it preserves non-JSON failure
bodies and injects status_code/request_id/is_error.
:meth:stream has no raw -- a streaming body is never
materialized.
Source code in dashscope/resources/audio/http_tts.py
693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 | |
stream
¶
stream(
text: str,
*,
model: str,
voice: str,
format: Optional[str] = None,
sample_rate: Optional[int] = None,
language_type: Optional[str] = None,
parameters: Optional[Dict[str, Any]] = None,
extra_input: Optional[Dict[str, Any]] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None
) -> HttpSynthesisStream
Stream the synthesis over SSE, one chunk per part.
Audio frames arrive base64-decoded on :attr:HttpSynthesisChunk.audio;
the final chunk (is_final) carries the whole-audio URL reference.
The envelope's raw HTTP headers are on
:attr:HttpSynthesisStream.headers.
Source code in dashscope/resources/audio/http_tts.py
759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 | |
AsyncHttpTts
¶
AsyncHttpTts(client)
Async HTTP TTS using the AsyncClient's own connection pool.
Source code in dashscope/resources/audio/http_tts.py
479 480 | |
synthesize
async
¶
synthesize(
text: str, **kwargs
) -> Union[HttpSynthesis, "APIResponse"]
Async twin of :meth:HttpTts.synthesize, raw included.
Source code in dashscope/resources/audio/http_tts.py
482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 | |
stream
async
¶
stream(text: str, **kwargs) -> AsyncHttpSynthesisStream
Source code in dashscope/resources/audio/http_tts.py
531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 | |
HttpSynthesis
¶
HttpSynthesisChunk
¶
Bases: ResponseBase
One SSE part of a streaming synthesis.
sentence
property
¶
sentence: Optional[Dict[str, Any]]
Sentence metadata of cosyvoice sentence-* parts.
HttpSynthesisStream
¶
HttpSynthesisStream(
inner: Generator[HttpSynthesisChunk, None, None],
response: Any,
)
Typed iterator over streaming synthesis parts, mirroring AppStream.
Source code in dashscope/resources/audio/http_tts.py
428 429 430 431 432 433 434 | |
AsyncHttpSynthesisStream
¶
AsyncHttpSynthesisStream(
inner: AsyncIterator[HttpSynthesisChunk], response
)
Async typed iterator over HTTP synthesis SSE parts.
Source code in dashscope/resources/audio/http_tts.py
457 458 459 460 461 462 463 | |
Recognition (ASR)¶
Asr
¶
Asr(client)
client.audio.asr: batch transcription + realtime recognition.
Source code in dashscope/resources/audio/asr.py
771 772 773 | |
transcribe
¶
transcribe(
file_urls: List[str],
*,
model: str = "qwen-audio-3.1-asr-flash-filetrans",
parameters: Optional[Dict[str, Any]] = None,
phrase_id: Optional[str] = None,
timeout: float = DEFAULT_WAIT_TIMEOUT,
submit_timeout: Optional[float] = None,
poll_interval: float = DEFAULT_POLL_INTERVAL,
max_interval: float = DEFAULT_MAX_INTERVAL,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None
) -> Job
Submit and wait; output.results carries per-file transcripts.
timeout is the wait budget — the same timeout as
jobs.wait. submit_timeout caps only the submit request.
Source code in dashscope/resources/audio/asr.py
775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 | |
recognize
¶
recognize(
*,
model: str,
format: str = "pcm",
sample_rate: int = 16000,
vocabulary_id: Optional[str] = None,
disfluency_removal_enabled: Optional[bool] = None,
diarization_enabled: Optional[bool] = None,
parameters: Optional[Dict[str, Any]] = None,
input: Optional[Dict[str, Any]] = None,
phrase_id: Optional[str] = None,
ws_url: Optional[str] = None,
connector: Optional[Connector] = None,
connection: Optional[WsConnection] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None
) -> Recognition
Open recognition; send audio and consume events concurrently.
With connection= the session runs as one task on the shared
socket (binary uplink is single-owner per connection, so one
active recognition per connection).
Source code in dashscope/resources/audio/asr.py
824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 | |
recognize_file
¶
recognize_file(
path: str,
*,
model: str,
format: str = "pcm",
sample_rate: int = 16000,
vocabulary_id: Optional[str] = None,
disfluency_removal_enabled: Optional[bool] = None,
diarization_enabled: Optional[bool] = None,
parameters: Optional[Dict[str, Any]] = None,
ws_url: Optional[str] = None,
connector: Optional[Connector] = None,
timeout: Optional[float] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None
) -> Iterator[Dict[str, Any]]
Recognize a local audio file, yielding the same events as
Recognition.events().
The file is auto-chunked and driven to completion in one session,
the 1.x Recognition.call(file) equivalent. Validation happens
eagerly; the session itself opens on first iteration.
Source code in dashscope/resources/audio/asr.py
870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 | |
AsyncAsr
¶
AsyncAsr(client)
Async client.audio.asr.
Realtime recognize sessions stay sync-only (design doc M3 3.5).
Source code in dashscope/resources/audio/asr.py
921 922 | |
transcribe
async
¶
transcribe(
file_urls: List[str],
*,
model: str = "qwen-audio-3.1-asr-flash-filetrans",
parameters: Optional[Dict[str, Any]] = None,
phrase_id: Optional[str] = None,
timeout: float = DEFAULT_WAIT_TIMEOUT,
submit_timeout: Optional[float] = None,
poll_interval: float = DEFAULT_POLL_INTERVAL,
max_interval: float = DEFAULT_MAX_INTERVAL,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None
) -> Job
Submit and wait; output.results carries per-file transcripts.
timeout is the wait budget — the same timeout as
jobs.wait. submit_timeout caps only the submit request.
Source code in dashscope/resources/audio/asr.py
924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 | |
Recognition
¶
Recognition(
client,
*,
model: str,
format: str = "pcm",
sample_rate: int = 16000,
vocabulary_id: Optional[str] = None,
disfluency_removal_enabled: Optional[bool] = None,
diarization_enabled: Optional[bool] = None,
parameters: Optional[Dict[str, Any]] = None,
input: Optional[Dict[str, Any]] = None,
phrase_id: Optional[str] = None,
ws_url: Optional[str] = None,
connector: Optional[Connector] = None,
connection: Optional[WsConnection] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None
)
Duplex recognition; consume events on one thread while sending audio.
Source code in dashscope/resources/audio/asr.py
568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 | |
start
¶
start() -> 'Recognition'
Source code in dashscope/resources/audio/asr.py
644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 | |
send_audio_frame
¶
send_audio_frame(data: bytes) -> None
Source code in dashscope/resources/audio/asr.py
682 683 684 685 | |
update_context
¶
update_context(context: Dict[str, Any]) -> None
Apply context to subsequent audio without restarting recognition.
Source code in dashscope/resources/audio/asr.py
687 688 689 690 691 | |
stop
¶
stop() -> None
End input; keep reading events through task-finished.
Source code in dashscope/resources/audio/asr.py
693 694 695 696 697 | |
events
¶
events() -> Iterator[Dict[str, Any]]
Source code in dashscope/resources/audio/asr.py
699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 | |
close
¶
close() -> None
Source code in dashscope/resources/audio/asr.py
735 736 737 738 739 740 | |
results
¶
results() -> Iterator[RecognitionResult]
Yield typed results until the task finishes.
Non-result events (task-finished, ...) are skipped: 1.x reports
them through on_complete instead of on_event. Heartbeat
sentences are dropped too, as 1.x does before invoking callbacks.
Source code in dashscope/resources/audio/asr.py
742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 | |
TranscriptionJobs
¶
TranscriptionJobs(client)
Async task facade over the batch transcription service.
Source code in dashscope/resources/audio/asr.py
126 127 128 | |
create
¶
create(
*,
model: str,
file_urls: List[str],
parameters: Optional[Dict[str, Any]] = None,
phrase_id: Optional[str] = None,
timeout: Optional[float] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[False] = False
) -> Job
create(
*,
model: str,
file_urls: List[str],
parameters: Optional[Dict[str, Any]] = None,
phrase_id: Optional[str] = None,
timeout: Optional[float] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[True]
) -> "APIResponse"
create(
*,
model: str,
file_urls: List[str],
parameters: Optional[Dict[str, Any]] = None,
phrase_id: Optional[str] = None,
timeout: Optional[float] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: bool = False
) -> Union[Job, "APIResponse"]
Submit a batch transcription task over one or more audio files.
raw=True returns the :class:~dashscope.transport.APIResponse
of the underlying call before any error check.
Source code in dashscope/resources/audio/asr.py
168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 | |
fetch
¶
fetch(
task_id: str,
*,
timeout: Optional[float] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[False] = False
) -> Job
fetch(
task_id: str,
*,
timeout: Optional[float] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[True]
) -> "APIResponse"
fetch(
task_id: str,
*,
timeout: Optional[float] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: bool = False
) -> Union[Job, "APIResponse"]
Fetch one task snapshot.
raw=True returns the :class:~dashscope.transport.APIResponse
of the underlying call before any error check.
Source code in dashscope/resources/audio/asr.py
236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 | |
wait
¶
wait(
job: Union[Job, str],
*,
poll_interval: float = DEFAULT_POLL_INTERVAL,
max_interval: float = DEFAULT_MAX_INTERVAL,
timeout: float = DEFAULT_WAIT_TIMEOUT,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
_sleep: Optional[Callable[[float], Any]] = None
) -> Job
Poll with backoff, without starting another poll after the deadline.
Source code in dashscope/resources/audio/asr.py
266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 | |
AsyncTranscriptionJobs
¶
AsyncTranscriptionJobs(client)
Source code in dashscope/resources/audio/asr.py
330 331 332 | |
create
async
¶
create(
*,
model: str,
file_urls: List[str],
parameters: Optional[Dict[str, Any]] = None,
phrase_id: Optional[str] = None,
timeout: Optional[float] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[False] = False
) -> Job
create(
*,
model: str,
file_urls: List[str],
parameters: Optional[Dict[str, Any]] = None,
phrase_id: Optional[str] = None,
timeout: Optional[float] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[True]
) -> "APIResponse"
create(
*,
model: str,
file_urls: List[str],
parameters: Optional[Dict[str, Any]] = None,
phrase_id: Optional[str] = None,
timeout: Optional[float] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: bool = False
) -> Union[Job, "APIResponse"]
Submit a batch transcription task over one or more audio files.
raw=True returns the :class:~dashscope.transport.APIResponse
of the underlying call before any error check.
Source code in dashscope/resources/audio/asr.py
372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 | |
fetch
async
¶
fetch(
task_id: str,
*,
timeout: Optional[float] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[False] = False
) -> Job
fetch(
task_id: str,
*,
timeout: Optional[float] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[True]
) -> "APIResponse"
fetch(
task_id: str,
*,
timeout: Optional[float] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: bool = False
) -> Union[Job, "APIResponse"]
Fetch one task snapshot.
raw=True returns the :class:~dashscope.transport.APIResponse
of the underlying call before any error check.
Source code in dashscope/resources/audio/asr.py
440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 | |
wait
async
¶
wait(
job: Union[Job, str],
*,
poll_interval: float = DEFAULT_POLL_INTERVAL,
max_interval: float = DEFAULT_MAX_INTERVAL,
timeout: float = DEFAULT_WAIT_TIMEOUT,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
_sleep: Optional[
Callable[[float], Awaitable[None]]
] = None
) -> Job
Async twin of :meth:TranscriptionJobs.wait.
Source code in dashscope/resources/audio/asr.py
470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 | |
RecognitionResult
¶
Bases: BaseModel
Typed view over one result-generated event payload.
Recognition.events() keeps yielding raw dicts; this model adds the
1.x RecognitionResult accessors (recognition.py:99-125) for
callers that want them. Unknown payload fields stay readable as
attributes (extra="allow").
get_sentence
¶
get_sentence() -> Optional[Dict[str, Any]]
The sentence dict of this result, or None.
Source code in dashscope/resources/audio/asr.py
548 549 550 551 552 | |
is_sentence_end
¶
is_sentence_end() -> bool
True when the sentence carries its end marker (end_time).
Source code in dashscope/resources/audio/asr.py
554 555 556 557 | |
get_usage
¶
get_usage() -> Optional[Dict[str, Any]]
Source code in dashscope/resources/audio/asr.py
559 560 561 562 | |
Realtime & Qwen Live¶
See the Qwen Live guide for a full
walkthrough of client.audio.live.
Realtime
¶
Realtime(client)
client.audio.realtime: Omni, Qwen-Audio and Livetranslate.
Source code in dashscope/resources/audio/realtime.py
886 887 | |
connect
¶
connect(
*,
model: str,
ws_url: Optional[str] = None,
connector: Optional[Connector] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None
) -> RealtimeSession
Create a session; enter its context or call session.connect().
Source code in dashscope/resources/audio/realtime.py
889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 | |
RealtimeSession
¶
RealtimeSession(
client,
*,
model: str,
ws_url: Optional[str] = None,
connector: Optional[Connector] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None
)
Bases: WsSessionBase
One realtime websocket session (context manager).
Usage::
with client.audio.realtime.connect(
model="qwen3.8-omni-flash-realtime",
) as rt:
rt.send_event({"type": "session.update", "session": {...}})
for event in rt.events():
...
Unlike the run-task family (asr/tts/tingwu), the
realtime protocol has no task id: one socket IS one session and the
server closes it when the session ends, so connection reuse does
not apply here -- every session opens its own socket (see
WsConnection for the multiplexable family). Latency-sensitive
callers can shave the handshake by connecting ahead of use rather
than by pooling.
Source code in dashscope/resources/audio/realtime.py
75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 | |
connect
¶
connect() -> 'RealtimeSession'
Open the connection, either explicitly or on context entry.
Source code in dashscope/resources/audio/realtime.py
127 128 129 130 131 132 133 134 135 136 | |
send_event
¶
send_event(event: Dict[str, Any]) -> None
Send a raw event, assigning an event_id unless one is supplied.
Source code in dashscope/resources/audio/realtime.py
166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 | |
update_session
¶
update_session(
output_modalities: List[str] = cast(List[str], _UNSET),
*,
voice: Optional[str] = cast(Optional[str], _UNSET),
input_audio_format: str = cast(str, _UNSET),
output_audio_format: str = cast(str, _UNSET),
enable_input_audio_transcription: bool = cast(
bool, _UNSET
),
input_audio_transcription_model: Optional[str] = cast(
Optional[str], _UNSET
),
enable_turn_detection: bool = cast(bool, _UNSET),
turn_detection_type: str = cast(str, _UNSET),
prefix_padding_ms: int = cast(int, _UNSET),
turn_detection_threshold: float = cast(float, _UNSET),
turn_detection_silence_duration_ms: int = cast(
int, _UNSET
),
turn_detection_param: Optional[Dict[str, Any]] = cast(
Optional[Dict[str, Any]], _UNSET
),
translation_params: Optional[Dict[str, Any]] = cast(
Optional[Dict[str, Any]], _UNSET
),
transcription_params: Optional[Dict[str, Any]] = cast(
Optional[Dict[str, Any]], _UNSET
),
input_audio_config: Optional[Dict[str, Any]] = cast(
Optional[Dict[str, Any]], _UNSET
),
output_audio_config: Optional[Dict[str, Any]] = cast(
Optional[Dict[str, Any]], _UNSET
),
input_video_representation_compact: Optional[
str
] = cast(Optional[str], _UNSET),
**kwargs: Any
) -> None
Patch modern sessions; pre-3.8 Omni retains its legacy defaults.
**kwargs is the protocol-evolution escape hatch: extra keys
merge into the wire session object after the named helpers,
so on key overlap the kwargs value wins and falsey values are
preserved as sent.
Source code in dashscope/resources/audio/realtime.py
206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 | |
append_audio
¶
append_audio(audio_b64: Union[str, bytes]) -> None
Append base64 audio, or encode raw bytes before appending.
Source code in dashscope/resources/audio/realtime.py
554 555 556 557 558 559 560 561 | |
append_video
¶
append_video(video_b64: Union[str, bytes]) -> None
Append a base64 image frame, or encode raw image bytes.
Source code in dashscope/resources/audio/realtime.py
563 564 565 566 567 568 569 570 | |
commit
¶
commit() -> None
Commit the appended audio and video.
Source code in dashscope/resources/audio/realtime.py
572 573 574 | |
clear_appended_audio
¶
clear_appended_audio() -> None
Source code in dashscope/resources/audio/realtime.py
576 577 | |
create_item
¶
create_item(
item: Dict[str, Any],
*,
previous_item_id: Optional[str] = None
) -> None
Source code in dashscope/resources/audio/realtime.py
579 580 581 582 583 584 585 586 587 588 | |
retrieve_item
¶
retrieve_item(item_id: str) -> None
Source code in dashscope/resources/audio/realtime.py
590 591 592 593 594 595 596 | |
delete_item
¶
delete_item(item_id: str) -> None
Source code in dashscope/resources/audio/realtime.py
598 599 600 601 602 603 604 | |
create_response
¶
create_response(
modalities: Optional[List[str]] = cast(
Optional[List[str]], _UNSET
),
instructions: Optional[str] = cast(
Optional[str], _UNSET
),
**kwargs: Any
) -> None
Send response.create.
**kwargs merges extra keys into the wire response object
(e.g. temperature), applied after modalities/
instructions and preserved verbatim, including falsey values.
Source code in dashscope/resources/audio/realtime.py
606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 | |
cancel_response
¶
cancel_response() -> None
Source code in dashscope/resources/audio/realtime.py
633 634 | |
end_session
¶
end_session() -> None
Request finish for Omni/Live; Qwen-Audio must drain then close().
Source code in dashscope/resources/audio/realtime.py
636 637 638 639 640 641 642 643 | |
recv_event
¶
recv_event() -> Optional[Dict[str, Any]]
Receive a raw event; unfinished responses cannot end silently.
Source code in dashscope/resources/audio/realtime.py
646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 | |
events
¶
events() -> Iterator[Dict[str, Any]]
Yield through session.finished, or until a normal close.
Source code in dashscope/resources/audio/realtime.py
872 873 874 875 876 877 878 879 880 | |
Live
¶
Live(client)
client.audio.live: Qwen-Live delegated voice sessions.
Source code in dashscope/resources/audio/live.py
711 712 | |
connect
¶
connect(
*,
model: str,
ws_url: Optional[str] = None,
connector: Optional[Connector] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None
) -> LiveSession
Create an unopened session; model routes the URL, not the front-end model.
Source code in dashscope/resources/audio/live.py
714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 | |
LiveSession
¶
LiveSession(
client,
*,
model: str,
ws_url: Optional[str] = None,
connector: Optional[Connector] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None
)
Bases: WsSessionBase
Live session; send helpers return request IDs, not acknowledgements.
events() yields recoverable error events; recv_event() raises them while leaving the connection open. Fatal errors and unexpected disconnects raise SDK exceptions in both forms. close() cancels locally, without draining.
Source code in dashscope/resources/audio/_ws_session.py
50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 | |
connect
¶
connect() -> LiveSession
Open the shared transport, mapping only Live's connection errors.
Source code in dashscope/resources/audio/live.py
192 193 194 195 196 197 198 199 200 | |
send_event
¶
send_event(event: Dict[str, Any]) -> str
Send a raw event and return its supplied or generated event_id.
Source code in dashscope/resources/audio/live.py
202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 | |
recv_event
¶
recv_event() -> Optional[Dict[str, Any]]
Receive an event; recoverable errors raise but leave the socket open.
Source code in dashscope/resources/audio/live.py
247 248 249 | |
events
¶
events() -> Iterator[Dict[str, Any]]
Yield recoverable errors; fatal failures raise, never becoming EOF.
Once started, an iterator belongs to that connection. Reconnecting requires a new iterator; audio and tasks are never replayed.
Source code in dashscope/resources/audio/live.py
407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 | |
cancel
¶
cancel() -> None
Immediately abort the socket without a WebSocket close handshake.
Source code in dashscope/resources/audio/live.py
426 427 428 | |
update_session
¶
update_session(
*,
model: str,
delegation: Dict[str, Any],
instructions: str = cast(str, _UNSET),
voice: str = cast(str, _UNSET),
tts: Dict[str, Any] = cast(Dict[str, Any], _UNSET),
input_audio_format: str = cast(str, _UNSET),
output_audio_format: str = cast(str, _UNSET),
tools: List[Dict[str, Any]] = cast(
List[Dict[str, Any]], _UNSET
),
transport: Dict[str, Any] = cast(
Dict[str, Any], _UNSET
),
modalities: List[str] = cast(List[str], _UNSET),
context: Dict[str, Any] = cast(Dict[str, Any], _UNSET),
turn_detection: Optional[Dict[str, Any]] = cast(
Optional[Dict[str, Any]], _UNSET
),
input_audio_transcription: Optional[
Dict[str, Any]
] = cast(Optional[Dict[str, Any]], _UNSET),
parameters: Optional[Dict[str, Any]] = None,
event_id: Optional[str] = None
) -> str
Configure the front end and delegation; confirmed by session.updated.
parameters merges extra keys into the session node for server
fields this SDK does not type yet; it may not shadow a typed argument.
Source code in dashscope/resources/audio/live.py
435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 | |
append_audio
¶
append_audio(
audio_b64: Union[str, bytes],
*,
event_id: Optional[str] = None
) -> str
Send raw or base64 16 kHz mono PCM16LE; AOQ input uses its own SDK.
Source code in dashscope/resources/audio/live.py
521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 | |
mute_input_audio
¶
mute_input_audio(*, event_id: Optional[str] = None) -> str
Replace subsequent input with silence without stopping playback.
Source code in dashscope/resources/audio/live.py
543 544 545 | |
unmute_input_audio
¶
unmute_input_audio(
*, event_id: Optional[str] = None
) -> str
Resume normal input without replaying muted audio.
Source code in dashscope/resources/audio/live.py
547 548 549 | |
append_instruction
¶
append_instruction(
content: str,
*,
delegation_id: str,
task_id: Optional[str] = None,
start_ms: Optional[float] = None,
end_ms: Optional[float] = None,
event_id: Optional[str] = None
) -> str
Yield for user input; acknowledged by session.instructions.appended.
Source code in dashscope/resources/audio/live.py
583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 | |
append_thinking
¶
append_thinking(
content: str,
*,
delegation_id: str,
start_ms: Optional[float] = None,
end_ms: Optional[float] = None,
event_id: Optional[str] = None
) -> str
Report silent client-task progress without ending the execution.
Source code in dashscope/resources/audio/live.py
604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 | |
append_commentary
¶
append_commentary(
content: str = cast(str, _UNSET),
*,
delegation_id: Optional[str] = None,
start_ms: Optional[float] = None,
end_ms: Optional[float] = None,
error: Optional[Dict[str, Any]] = None,
event_id: Optional[str] = None
) -> str
Send user text or a client-task result; content may be omitted.
Source code in dashscope/resources/audio/live.py
624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 | |
create_session_item
¶
create_session_item(
item: Dict[str, Any] = cast(Dict[str, Any], _UNSET),
*,
event_id: Optional[str] = None
) -> str
Send session.item.create with an optional front-end item.
Source code in dashscope/resources/audio/live.py
646 647 648 649 650 651 652 653 654 655 656 | |
submit_function_output
¶
submit_function_output(
call_id: str,
output: Any,
*,
event_id: Optional[str] = None
) -> str
Reply to a front-end function from session.event, never a backend tool.
Source code in dashscope/resources/audio/live.py
658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 | |
create_response_item
¶
create_response_item(
item: Dict[str, Any], *, event_id: Optional[str] = None
) -> str
Buffer next-task input in Responses mode, not an active task's result.
Source code in dashscope/resources/audio/live.py
679 680 681 682 683 684 685 686 687 688 689 | |
create_response
¶
create_response(
response: Optional[Dict[str, Any]] = None,
*,
event_id: Optional[str] = None
) -> str
Start a new Responses task, or request a reply from a capable front end.
Source code in dashscope/resources/audio/live.py
691 692 693 694 695 696 697 698 699 700 701 | |
end_session
¶
end_session(*, event_id: Optional[str] = None) -> str
Send session.close; keep receiving until session.closed or disconnect.
Source code in dashscope/resources/audio/live.py
703 704 705 | |
Voice & vocabulary customization¶
Voices
¶
Voices(client)
Bases: _CustomizationBase
client.audio.voices -- voice cloning.
Source code in dashscope/resources/audio/customization.py
130 131 | |
create
¶
create(
*,
target_model: str,
prefix: str,
url: Optional[str] = None,
audio: Optional[str] = None,
language_hints: Optional[List[str]] = None,
max_prompt_audio_length: Optional[float] = None,
extra_input: Optional[Dict[str, Any]] = None,
model: Optional[str] = None,
parameters: Optional[Dict[str, Any]] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[False] = False
) -> Voice
create(
*,
target_model: str,
prefix: str,
url: Optional[str] = None,
audio: Optional[str] = None,
language_hints: Optional[List[str]] = None,
max_prompt_audio_length: Optional[float] = None,
extra_input: Optional[Dict[str, Any]] = None,
model: Optional[str] = None,
parameters: Optional[Dict[str, Any]] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[True]
) -> "APIResponse"
create(
*,
target_model: str,
prefix: str,
url: Optional[str] = None,
audio: Optional[str] = None,
language_hints: Optional[List[str]] = None,
max_prompt_audio_length: Optional[float] = None,
extra_input: Optional[Dict[str, Any]] = None,
model: Optional[str] = None,
parameters: Optional[Dict[str, Any]] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: bool = False
) -> Union[Voice, "APIResponse"]
Clone a voice from audio (Data URI/URL) or a legacy url.
Provide exactly one source. The result carries voice_id.
raw=True returns the :class:~dashscope.transport.APIResponse
before any error check, so a caller can read the status, headers and
elapsed of the response or re-parse it itself. The non-raw path
keeps response_body's envelope: it preserves non-JSON failure
bodies and injects status_code/request_id/is_error.
Source code in dashscope/resources/audio/customization.py
686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 | |
list
¶
list(
*,
prefix: Optional[str] = None,
page_index: int = 0,
page_size: int = 10,
model: Optional[str] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[False] = False
) -> VoiceList
list(
*,
prefix: Optional[str] = None,
page_index: int = 0,
page_size: int = 10,
model: Optional[str] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[True]
) -> "APIResponse"
list(
*,
prefix: Optional[str] = None,
page_index: int = 0,
page_size: int = 10,
model: Optional[str] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: bool = False
) -> Union[VoiceList, "APIResponse"]
List cloned voices, optionally narrowed to one prefix.
raw=True returns the :class:~dashscope.transport.APIResponse
before any error check, so a caller can read the status, headers and
elapsed of the response or re-parse it itself. The non-raw path
keeps response_body's envelope: it preserves non-JSON failure
bodies and injects status_code/request_id/is_error.
Source code in dashscope/resources/audio/customization.py
798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 | |
get
¶
get(
voice_id: str,
*,
model: Optional[str] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[False] = False
) -> Voice
get(
voice_id: str,
*,
model: Optional[str] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[True]
) -> "APIResponse"
get(
voice_id: str,
*,
model: Optional[str] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: bool = False
) -> Union[Voice, "APIResponse"]
Fetch one voice's details (1.x query_voice).
raw=True returns the :class:~dashscope.transport.APIResponse
before any error check, so a caller can read the status, headers and
elapsed of the response or re-parse it itself. The non-raw path
keeps response_body's envelope: it preserves non-JSON failure
bodies and injects status_code/request_id/is_error.
Source code in dashscope/resources/audio/customization.py
885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 | |
update
¶
update(
voice_id: str,
*,
url: str,
model: Optional[str] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[False] = False
) -> JobStatus
update(
voice_id: str,
*,
url: str,
model: Optional[str] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[True]
) -> "APIResponse"
update(
voice_id: str,
*,
url: str,
model: Optional[str] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: bool = False
) -> Union[JobStatus, "APIResponse"]
Re-clone one voice from the audio at url.
raw=True returns the :class:~dashscope.transport.APIResponse
before any error check, so a caller can read the status, headers and
elapsed of the response or re-parse it itself. The non-raw path
keeps response_body's envelope: it preserves non-JSON failure
bodies and injects status_code/request_id/is_error.
Source code in dashscope/resources/audio/customization.py
967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 | |
delete
¶
delete(
voice_id: str,
*,
model: Optional[str] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[False] = False
) -> JobStatus
delete(
voice_id: str,
*,
model: Optional[str] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[True]
) -> "APIResponse"
delete(
voice_id: str,
*,
model: Optional[str] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: bool = False
) -> Union[JobStatus, "APIResponse"]
Delete one cloned voice.
raw=True returns the :class:~dashscope.transport.APIResponse
before any error check, so a caller can read the status, headers and
elapsed of the response or re-parse it itself. The non-raw path
keeps response_body's envelope: it preserves non-JSON failure
bodies and injects status_code/request_id/is_error.
Source code in dashscope/resources/audio/customization.py
1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 | |
Voice
¶
Bases: ResponseBase
One cloned voice; create/get return it.
VoiceList
¶
Bases: ResponseBase
voices
class-attribute
instance-attribute
¶
voices: List[Dict[str, Any]] = Field(
default_factory=list,
validation_alias=AliasChoices("voices", "voice_list"),
)
VoiceDesign
¶
VoiceDesign(client)
Bases: _CustomizationBase
client.audio.voice_design -- voices described in words, not audio.
Shares the voice-enrollment registry with :class:Voices: same path,
same model, same action words, same voice_id, one quota. Only
create differs -- a text prompt where cloning has a sample url
-- so list returns cloned voices as well as designed ones.
Source code in dashscope/resources/audio/customization.py
130 131 | |
create
¶
create(
*,
target_model: str,
prefix: str,
voice_prompt: str,
preview_text: str,
language_hints: Optional[List[str]] = None,
extra_input: Optional[Dict[str, Any]] = None,
model: Optional[str] = None,
parameters: Optional[Dict[str, Any]] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[False] = False
) -> DesignedVoice
create(
*,
target_model: str,
prefix: str,
voice_prompt: str,
preview_text: str,
language_hints: Optional[List[str]] = None,
extra_input: Optional[Dict[str, Any]] = None,
model: Optional[str] = None,
parameters: Optional[Dict[str, Any]] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[True]
) -> "APIResponse"
create(
*,
target_model: str,
prefix: str,
voice_prompt: str,
preview_text: str,
language_hints: Optional[List[str]] = None,
extra_input: Optional[Dict[str, Any]] = None,
model: Optional[str] = None,
parameters: Optional[Dict[str, Any]] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: bool = False
) -> Union[DesignedVoice, "APIResponse"]
Design a voice from a text prompt; carries voice_id + preview.
prefix becomes part of the generated id,
{target_model}-vd-{prefix}-{unique}. Only the first element of
language_hints is honoured, and it must match the language of
preview_text. The voice is unusable until its status leaves
DEPLOYING.
raw=True returns the :class:~dashscope.transport.APIResponse
before any error check, so a caller can read the status, headers and
elapsed of the response or re-parse it itself. The non-raw path
keeps response_body's envelope: it preserves non-JSON failure
bodies and injects status_code/request_id/is_error.
Source code in dashscope/resources/audio/customization.py
1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 | |
list
¶
list(
*,
prefix: Optional[str] = None,
page_index: int = 0,
page_size: int = 10,
model: Optional[str] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[False] = False
) -> DesignedVoiceList
list(
*,
prefix: Optional[str] = None,
page_index: int = 0,
page_size: int = 10,
model: Optional[str] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[True]
) -> "APIResponse"
list(
*,
prefix: Optional[str] = None,
page_index: int = 0,
page_size: int = 10,
model: Optional[str] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: bool = False
) -> Union[DesignedVoiceList, "APIResponse"]
List voices, optionally narrowed to one prefix.
raw=True returns the :class:~dashscope.transport.APIResponse
before any error check, so a caller can read the status, headers and
elapsed of the response or re-parse it itself. The non-raw path
keeps response_body's envelope: it preserves non-JSON failure
bodies and injects status_code/request_id/is_error.
Source code in dashscope/resources/audio/customization.py
1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 | |
get
¶
get(
voice_id: str,
*,
model: Optional[str] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[False] = False
) -> DesignedVoice
get(
voice_id: str,
*,
model: Optional[str] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[True]
) -> "APIResponse"
get(
voice_id: str,
*,
model: Optional[str] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: bool = False
) -> Union[DesignedVoice, "APIResponse"]
Fetch one voice's details, including its review status.
raw=True returns the :class:~dashscope.transport.APIResponse
before any error check, so a caller can read the status, headers and
elapsed of the response or re-parse it itself. The non-raw path
keeps response_body's envelope: it preserves non-JSON failure
bodies and injects status_code/request_id/is_error.
Source code in dashscope/resources/audio/customization.py
1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 | |
delete
¶
delete(
voice_id: str,
*,
model: Optional[str] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[False] = False
) -> JobStatus
delete(
voice_id: str,
*,
model: Optional[str] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[True]
) -> "APIResponse"
delete(
voice_id: str,
*,
model: Optional[str] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: bool = False
) -> Union[JobStatus, "APIResponse"]
Delete one designed voice, freeing its quota.
raw=True returns the :class:~dashscope.transport.APIResponse
before any error check, so a caller can read the status, headers and
elapsed of the response or re-parse it itself. The non-raw path
keeps response_body's envelope: it preserves non-JSON failure
bodies and injects status_code/request_id/is_error.
Source code in dashscope/resources/audio/customization.py
1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 | |
DesignedVoice
¶
Bases: ResponseBase
One designed voice.
The same registry entry as a cloned :class:Voice -- one endpoint, one
voice_id, one shared quota -- plus the two texts it was designed
from.
DesignedVoiceList
¶
Bases: ResponseBase
voices
class-attribute
instance-attribute
¶
voices: List[Dict[str, Any]] = Field(
default_factory=list,
validation_alias=AliasChoices("voices", "voice_list"),
)
Vocabularies
¶
Vocabularies(client)
Bases: _CustomizationBase
client.audio.vocabularies -- asr hot-word tables.
Source code in dashscope/resources/audio/customization.py
130 131 | |
create
¶
create(
*,
target_model: str,
prefix: str,
vocabulary: List[Dict[str, Any]],
model: Optional[str] = None,
parameters: Optional[Dict[str, Any]] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[False] = False
) -> Vocabulary
create(
*,
target_model: str,
prefix: str,
vocabulary: List[Dict[str, Any]],
model: Optional[str] = None,
parameters: Optional[Dict[str, Any]] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[True]
) -> "APIResponse"
create(
*,
target_model: str,
prefix: str,
vocabulary: List[Dict[str, Any]],
model: Optional[str] = None,
parameters: Optional[Dict[str, Any]] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: bool = False
) -> Union[Vocabulary, "APIResponse"]
Create a table; the result carries vocabulary_id.
raw=True returns the :class:~dashscope.transport.APIResponse
before any error check, so a caller can read the status, headers and
elapsed of the response or re-parse it itself. The non-raw path
keeps response_body's envelope: it preserves non-JSON failure
bodies and injects status_code/request_id/is_error.
Source code in dashscope/resources/audio/customization.py
244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 | |
list
¶
list(
*,
prefix: Optional[str] = None,
page_index: int = 0,
page_size: int = 10,
model: Optional[str] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[False] = False
) -> VocabularyList
list(
*,
prefix: Optional[str] = None,
page_index: int = 0,
page_size: int = 10,
model: Optional[str] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[True]
) -> "APIResponse"
list(
*,
prefix: Optional[str] = None,
page_index: int = 0,
page_size: int = 10,
model: Optional[str] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: bool = False
) -> Union[VocabularyList, "APIResponse"]
List tables, optionally narrowed to one prefix.
raw=True returns the :class:~dashscope.transport.APIResponse
before any error check, so a caller can read the status, headers and
elapsed of the response or re-parse it itself. The non-raw path
keeps response_body's envelope: it preserves non-JSON failure
bodies and injects status_code/request_id/is_error.
Source code in dashscope/resources/audio/customization.py
335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 | |
get
¶
get(
vocabulary_id: str,
*,
model: Optional[str] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[False] = False
) -> Vocabulary
get(
vocabulary_id: str,
*,
model: Optional[str] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[True]
) -> "APIResponse"
get(
vocabulary_id: str,
*,
model: Optional[str] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: bool = False
) -> Union[Vocabulary, "APIResponse"]
Fetch one table's contents (1.x query_vocabulary).
raw=True returns the :class:~dashscope.transport.APIResponse
before any error check, so a caller can read the status, headers and
elapsed of the response or re-parse it itself. The non-raw path
keeps response_body's envelope: it preserves non-JSON failure
bodies and injects status_code/request_id/is_error.
Source code in dashscope/resources/audio/customization.py
422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 | |
update
¶
update(
vocabulary_id: str,
*,
vocabulary: List[Dict[str, Any]],
model: Optional[str] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[False] = False
) -> JobStatus
update(
vocabulary_id: str,
*,
vocabulary: List[Dict[str, Any]],
model: Optional[str] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[True]
) -> "APIResponse"
update(
vocabulary_id: str,
*,
vocabulary: List[Dict[str, Any]],
model: Optional[str] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: bool = False
) -> Union[JobStatus, "APIResponse"]
Replace one table's contents.
raw=True returns the :class:~dashscope.transport.APIResponse
before any error check, so a caller can read the status, headers and
elapsed of the response or re-parse it itself. The non-raw path
keeps response_body's envelope: it preserves non-JSON failure
bodies and injects status_code/request_id/is_error.
Source code in dashscope/resources/audio/customization.py
504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 | |
delete
¶
delete(
vocabulary_id: str,
*,
model: Optional[str] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[False] = False
) -> JobStatus
delete(
vocabulary_id: str,
*,
model: Optional[str] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[True]
) -> "APIResponse"
delete(
vocabulary_id: str,
*,
model: Optional[str] = None,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: bool = False
) -> Union[JobStatus, "APIResponse"]
Delete one table.
raw=True returns the :class:~dashscope.transport.APIResponse
before any error check, so a caller can read the status, headers and
elapsed of the response or re-parse it itself. The non-raw path
keeps response_body's envelope: it preserves non-JSON failure
bodies and injects status_code/request_id/is_error.
Source code in dashscope/resources/audio/customization.py
585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 | |
Vocabulary
¶
Bases: ResponseBase
One hot-word table; create/get return it.
VocabularyList
¶
Bases: ResponseBase
vocabularies
class-attribute
instance-attribute
¶
vocabularies: List[Dict[str, Any]] = Field(
default_factory=list,
validation_alias=AliasChoices(
"vocabularies", "vocabulary_list"
),
)
Phrases
¶
Phrases(client)
client.audio.phrases -- paraformer hot phrases.
Source code in dashscope/resources/audio/phrases.py
119 120 | |
create
¶
create(
*,
model: str,
phrases: Dict[str, Any],
training_type: str = "compile_asr_phrase",
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[False] = False
) -> Phrase
create(
*,
model: str,
phrases: Dict[str, Any],
training_type: str = "compile_asr_phrase",
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[True]
) -> "APIResponse"
create(
*,
model: str,
phrases: Dict[str, Any],
training_type: str = "compile_asr_phrase",
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: bool = False
) -> Union[Phrase, "APIResponse"]
Compile a new phrase set; the result carries phrase_id.
raw=True returns the :class:~dashscope.transport.APIResponse
before any error check, so a caller can read the status, headers and
elapsed of the response or re-parse it itself. The non-raw path
keeps response_body's envelope: it preserves non-JSON failure
bodies and injects status_code/request_id/is_error.
Source code in dashscope/resources/audio/phrases.py
169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 | |
update
¶
update(
phrase_id: str,
*,
model: str,
phrases: Dict[str, Any],
training_type: str = "compile_asr_phrase",
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[False] = False
) -> Phrase
update(
phrase_id: str,
*,
model: str,
phrases: Dict[str, Any],
training_type: str = "compile_asr_phrase",
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[True]
) -> "APIResponse"
update(
phrase_id: str,
*,
model: str,
phrases: Dict[str, Any],
training_type: str = "compile_asr_phrase",
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: bool = False
) -> Union[Phrase, "APIResponse"]
Recompile the phrase set behind phrase_id.
raw=True returns the :class:~dashscope.transport.APIResponse
before any error check, so a caller can read the status, headers and
elapsed of the response or re-parse it itself. The non-raw path
keeps response_body's envelope: it preserves non-JSON failure
bodies and injects status_code/request_id/is_error.
Source code in dashscope/resources/audio/phrases.py
267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 | |
get
¶
get(
phrase_id: str,
*,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[False] = False
) -> Phrase
get(
phrase_id: str,
*,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[True]
) -> "APIResponse"
get(
phrase_id: str,
*,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: bool = False
) -> Union[Phrase, "APIResponse"]
Fetch one compiled phrase set (1.x query_phrases).
raw=True returns the :class:~dashscope.transport.APIResponse
before any error check, so a caller can read the status, headers and
elapsed of the response or re-parse it itself. The non-raw path
keeps response_body's envelope: it preserves non-JSON failure
bodies and injects status_code/request_id/is_error.
Source code in dashscope/resources/audio/phrases.py
361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 | |
list
¶
list(
*,
page_no: int = 1,
page_size: int = 10,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[False] = False
) -> PhraseList
list(
*,
page_no: int = 1,
page_size: int = 10,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[True]
) -> "APIResponse"
list(
*,
page_no: int = 1,
page_size: int = 10,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: bool = False
) -> Union[PhraseList, "APIResponse"]
List compiled phrase sets.
raw=True returns the :class:~dashscope.transport.APIResponse
before any error check, so a caller can read the status, headers and
elapsed of the response or re-parse it itself. The non-raw path
keeps response_body's envelope: it preserves non-JSON failure
bodies and injects status_code/request_id/is_error.
Source code in dashscope/resources/audio/phrases.py
447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 | |
delete
¶
delete(
phrase_id: str,
*,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[False] = False
) -> JobStatus
delete(
phrase_id: str,
*,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: Literal[True]
) -> "APIResponse"
delete(
phrase_id: str,
*,
timeout: Optional[float] = None,
headers: Optional[Mapping[str, str]] = None,
api_key: Optional[str] = None,
workspace: Optional[str] = None,
extra_headers: Optional[Mapping[str, str]] = None,
extra_query: Optional[Mapping[str, Any]] = None,
extra_body: Optional[Mapping[str, Any]] = None,
max_retries: Optional[int] = None,
raw: bool = False
) -> Union[JobStatus, "APIResponse"]
Delete one compiled phrase set.
raw=True returns the :class:~dashscope.transport.APIResponse
before any error check, so a caller can read the status, headers and
elapsed of the response or re-parse it itself. The non-raw path
keeps response_body's envelope: it preserves non-JSON failure
bodies and injects status_code/request_id/is_error.
Source code in dashscope/resources/audio/phrases.py
532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 | |
Phrase
¶
Bases: ResponseBase
One compiled phrase set; prefer its resource ID over the compile job ID.
PhraseList
¶
Bases: ResponseBase
total_count
class-attribute
instance-attribute
¶
total_count: int = Field(
0, validation_alias=AliasChoices("total_count", "total")
)
phrases
class-attribute
instance-attribute
¶
phrases: List[Phrase] = Field(
default_factory=list,
validation_alias=AliasChoices(
"phrases", "jobs", "outputs", "finetuned_outputs"
),
)