PR #25648: Allow segments and language in ovhcloud response in case of response_format=verbose_json requests

• Repository: BerriAI/litellm
• Author: marcelthole
• State: open | merged: False
• Link: https://github.com/BerriAI/litellm/pull/25648

Description (problem / solution / changelog)

Relevant issues

Fixes #25633

Pre-Submission checklist

Please complete all items before asking a LiteLLM maintainer to review your PR

• I have Added testing in the tests/test_litellm/ directory, Adding at least 1 test is a hard requirement - see details
• My PR passes all unit tests on make test-unit
• My PR's scope is as isolated as possible, it only solves 1 specific problem
• I have requested a Greptile review by commenting @greptileai and received a Confidence Score of at least 4/5 before requesting a maintainer review

Delays in PR merge?

If you're seeing a delay in your PR being merged, ping the LiteLLM Team on Slack (#pr-review).

CI (LiteLLM team)

CI status guideline:

• 50-55 passing tests: main is stable with minor issues.
• 45-49 passing tests: acceptable but needs attention
• <= 40 passing tests: unstable; be careful with your merges and assess the risk.

• Branch creation CI run
  Link:

• CI run for the last commit
  Link:

• Merge / cherry-pick CI run
  Links:

Screenshots / Proof of Fix

<!-- Include screenshots, screen recordings, or log output demonstrating that your changes work as expected. For bug fixes: show reproduction before the fix and passing behavior after. For new features: show the feature working end-to-end. For UI changes: include before/after screenshots. -->

Type

<!-- Select the type of Pull Request --> <!-- Keep only the necessary ones -->

🐛 Bug Fix ✅ Test

Changes

Changed files

• litellm/llms/ovhcloud/audio_transcription/transformation.py (modified, +5/-0)
• tests/test_litellm/llms/ovhcloud/test_ovhcloud_audio_transcription_transformation.py (modified, +38/-0)

---

PR #25651: fix(ovhcloud): preserve segments/words/language in verbose_json transcription response

• Repository: BerriAI/litellm
• Author: Anai-Guo
• State: open | merged: False
• Link: https://github.com/BerriAI/litellm/pull/25651

Description (problem / solution / changelog)

Problem

Fixes #25633

When using with the OVHCloud Whisper endpoint, the API returns a full verbose response containing segments, words, language, duration, etc. — but only the text field was being returned to callers.

Root cause: OVHCloudAudioTranscriptionConfig.transform_audio_transcription_response manually extracted only the text (or transcript) field, discarding everything else.

Fix

Mirror the approach used by the OpenAI Whisper implementation: pass the full response JSON to TranscriptionResponse(**response_json) so all OpenAI-compatible fields are preserved.

A normalisation step handles the case where OVHCloud returns transcript instead of text (backwards compatible).

Changes

• litellm/llms/ovhcloud/audio_transcription/transformation.py — updated transform_audio_transcription_response to pass through all fields
• tests/test_litellm/llms/ovhcloud/test_ovhcloud_audio_transcription_transformation.py — added unit tests covering:
  • verbose_json response with segments preserved
  • plain-text response (no regression)
  • transcript field normalised to text

Testing

The new unit tests run without an API key and cover the fixed code path directly.

Changed files

• litellm/llms/ovhcloud/audio_transcription/transformation.py (modified, +18/-2)
• tests/test_litellm/llms/ovhcloud/test_ovhcloud_audio_transcription_transformation.py (modified, +73/-0)