Update docs with ASRPipeline#4128
Open
as-suvorov wants to merge 4 commits into
Open
Conversation
Contributor
There was a problem hiding this comment.
Pull request overview
Note
Copilot couldn't run its full agentic review because no GitHub Actions runner was available. Make sure your repository has a runner available to run Copilot's review, or add a copilot-setup-steps.yml file specifying one with the runs-on attribute. See the docs for more details.
Updates Speech Recognition documentation and site snippets to use the newer ASR pipeline terminology/APIs (while keeping JS examples on Whisper where ASRPipeline isn’t available).
Changes:
- Renamed “Whisper” speech recognition docs/sections to “Automatic Speech Recognition” and updated copy/links accordingly.
- Updated Python/C++ snippets and API references from
WhisperPipeline/WhisperGenerationConfigtoASRPipeline/ASRGenerationConfig. - Added Qwen3-ASR models and export notes to the supported model documentation.
Reviewed changes
Copilot reviewed 11 out of 11 changed files in this pull request and generated 3 comments.
Show a summary per file
| File | Description |
|---|---|
| site/src/pages/_sections/UseCasesSection/components/speech-recognition.tsx | Updates use-case card title/description to generic ASR wording. |
| site/src/pages/_sections/HeroSection/PipelinesCarousel/index.tsx | Switches carousel code snippet to ASRPipeline. |
| site/docs/use-cases/speech-recognition/index.mdx | Renames page and adjusts model prep wording/links. |
| site/docs/use-cases/speech-recognition/_sections/_usage_options/index.mdx | Broad ASR doc refresh: pipeline names, config types, and Whisper-specific feature grouping. |
| site/docs/use-cases/speech-recognition/_sections/_run_model/index.mdx | Updates pipeline reference and input expectations text to ASR. |
| site/docs/use-cases/speech-recognition/_sections/_run_model/_code_example_python.mdx | Replaces WhisperPipeline with ASRPipeline in Python snippet. |
| site/docs/use-cases/speech-recognition/_sections/_run_model/_code_example_cpp.mdx | Updates include path and C++ snippet to ASRPipeline. |
| site/docs/supported-models/index.mdx | Renames section and adds Qwen3-ASR export note. |
| site/docs/supported-models/_components/whisper-models-table/models.ts | Adds Qwen3-ASR entries to the models list. |
| site/docs/guides/streaming.mdx | Updates streaming support list to ASRPipeline. |
| site/docs/getting-started/introduction.mdx | Updates intro text to reference ASR instead of Whisper-specific wording. |
Comment on lines
+397
to
399
| ov::genai::ASRPipeline pipe(model_path, "CPU"); | ||
|
|
||
| auto result = pipeline.generate(raw_speech); |
| Download and convert model (e.g. [openai/whisper-base](https://huggingface.co/openai/whisper-base)) to OpenVINO format from Hugging Face: | ||
| Download and convert an automatic speech recognition model (e.g. [openai/whisper-base](https://huggingface.co/openai/whisper-base)) to OpenVINO format from Hugging Face: | ||
|
|
||
| <Tabs groupId="whisper-export-command"> |
Comment on lines
9
to
+21
| export const WHISPER_MODELS: WhisperModelType[] = [ | ||
| { | ||
| architecture: 'Qwen3ASRForConditionalGeneration', | ||
| models: [ | ||
| { | ||
| name: 'Qwen3-ASR', | ||
| links: [ | ||
| 'https://huggingface.co/Qwen/Qwen3-ASR-0.6B', | ||
| 'https://huggingface.co/Qwen/Qwen3-ASR-1.7B', | ||
| ], | ||
| }, | ||
| ], | ||
| }, |
yatarkan
reviewed
Jul 10, 2026
| @@ -7,6 +7,18 @@ type WhisperModelType = { | |||
| }; | |||
|
|
|||
| export const WHISPER_MODELS: WhisperModelType[] = [ | |||
Contributor
There was a problem hiding this comment.
Should we also rename this?
Suggested change
| export const WHISPER_MODELS: WhisperModelType[] = [ | |
| export const SPEECH_RECOGNITION_MODELS: SpeechRecognitionModelType[] = [ |
or
Suggested change
| export const WHISPER_MODELS: WhisperModelType[] = [ | |
| export const ASR_MODELS: ASRModelType[] = [ |
| #### Transcription | ||
|
|
||
| Whisper models can automatically detect the language of the input audio, or you can specify the language to improve accuracy. The detected (or specified) language is always available in the result via the `language` field: | ||
| ASR models can automatically detect the language of the input audio, or you can specify the language to improve accuracy. The detected (or specified) language is available in the result via the `languages` field: |
Contributor
There was a problem hiding this comment.
If there is no abbreviation description in the page above (before this section).
Suggested change
| ASR models can automatically detect the language of the input audio, or you can specify the language to improve accuracy. The detected (or specified) language is available in the result via the `languages` field: | |
| Automatic speech recognition (ASR) models can automatically detect the language of the input audio, or you can specify the language to improve accuracy. The detected (or specified) language is available in the result via the `languages` field: |
Contributor
There was a problem hiding this comment.
Maybe it is better to have "Automatic speech recognition (ASR)" in the beginning of the page (or even in page header) so we can refer just "ASR models" further on the page.
| Beam search support depends on the selected model architecture. Qwen3-ASR models do not support beam search decoding. | ||
| ::: | ||
|
|
||
| #### Transcription |
Contributor
There was a problem hiding this comment.
Changing header L3 to L4 might exclude it from right side page navigation, please double check.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Description
JS bindings do not have ASRPipeline interface implemented. So I kept WhisperPipeline in js tabs.
Checklist: