Document how to use open asr leaderboard to benchmark ASR model accuracy#4355
Open
michalkulakowski wants to merge 3 commits into
Open
Document how to use open asr leaderboard to benchmark ASR model accuracy#4355michalkulakowski wants to merge 3 commits into
michalkulakowski wants to merge 3 commits into
Conversation
Contributor
There was a problem hiding this comment.
Pull request overview
This PR aims to document how to benchmark ASR model transcription accuracy using the Open ASR Leaderboard tooling within the OVMS audio demo documentation, but it also modifies build dependency sourcing for OpenVINO GenAI.
Changes:
- Added a new “Evaluate transcription accuracy with Open ASR Leaderboard” section to
demos/audio/README.md. - Updated
versions.mkto point OpenVINO GenAI source checkout to a different commit and a different GitHub organization.
Reviewed changes
Copilot reviewed 2 out of 2 changed files in this pull request and generated 2 comments.
| File | Description |
|---|---|
versions.mk |
Changes the pinned OpenVINO GenAI commit and switches the source org used by build scripts. |
demos/audio/README.md |
Adds step-by-step instructions for evaluating WER/CER with Open ASR Leaderboard. |
Comment on lines
+24
to
+28
| OV_GENAI_BRANCH ?= 27b480d39731996f68ed8cf7c54a5c0dd6122ba6 | ||
|
|
||
| # Source repository organizations | ||
| OV_SOURCE_ORG ?= openvinotoolkit | ||
| OV_GENAI_ORG ?= openvinotoolkit | ||
| OV_GENAI_ORG ?= as-suvorov |
Comment on lines
+372
to
+377
| 2. Apply OVMS API compatibility patch: | ||
| ```bash | ||
| git apply ../external/open_asr_leaderboard.patch | ||
| ``` | ||
|
|
||
| If you cloned the leaderboard repository in a different location, adjust the patch path accordingly. |
e2f66ae to
d449971
Compare
Comment on lines
+15
to
+22
| @@ -2,7 +2,7 @@ | ||
| from io import BytesIO | ||
| from typing import Optional | ||
|
|
||
| -import openai | ||
| +from openai import OpenAI | ||
|
|
||
| from . import APIProvider, register |
Comment on lines
+24
to
+43
| @@ -17,10 +17,11 @@ def transcribe( | ||
| use_url: bool = False, | ||
| language: str = "en", | ||
| ) -> str: | ||
| + client = OpenAI() | ||
| if use_url: | ||
| response = requests.get(sample["row"]["audio"][0]["src"]) | ||
| audio_data = BytesIO(response.content) | ||
| - response = openai.Audio.transcribe( | ||
| + response = client.audio.transcriptions.create( | ||
| model=model_variant, | ||
| file=audio_data, | ||
| response_format="text", | ||
| @@ -29,7 +30,7 @@ def transcribe( | ||
| ) | ||
| else: | ||
| with open(audio_file_path, "rb") as audio_file: | ||
| - response = openai.Audio.transcribe( | ||
| + response = client.audio.transcriptions.create( | ||
| model=model_variant, |
a379284 to
b1452cd
Compare
added 2 commits
July 3, 2026 11:46
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.
🛠 Summary
JIRA/Issue if applicable.
Describe the changes.
🧪 Checklist
``