MLE-5321: docs(audio): broaden supported file formats to include ogg, opus, aac

[rishabh-bhargava]

Summary

Adds .ogg, .opus, and .aac to the documented supported-format list for /audio/transcriptions and /audio/translations. The decoder is shared across STT models, so these formats were already supported in practice — just not in the docs.

Verification

End-to-end on api.together.ai, both endpoints, three models, three new formats — 9 combinations all return HTTP 200 with the correct LibriSpeech reference transcript:

Endpoint	Model	.ogg	.opus	.aac
/audio/transcriptions	nvidia/parakeet-tdt-0.6b-v3	✓	✓	✓
/audio/transcriptions	openai/whisper-large-v3	✓	✓	✓
/audio/transcriptions	mistralai/Voxtral-Mini-3B-2507	✓	✓	✓
/audio/translations	openai/whisper-large-v3	✓	✓	✓

Reproduction:

curl -sS -X POST https://api.together.ai/v1/audio/transcriptions \
  -H "Authorization: Bearer \$TOGETHER_API_KEY" \
  -F "model=nvidia/parakeet-tdt-0.6b-v3" \
  -F "file=@audio.opus"

Test plan

• OpenAPI YAML lints cleanly (no schema breakage — only description string changes)
• Mintlify replication picks up the description change automatically (no action needed in mintlify-docs/openapi.yaml per the team's replication setup)

How this was found

Surfaced by the formats.unsupported test in the new STT feature-test framework (benchmarks/stt/features/ on the together-voice repo). The test originally probed for a 4xx on .ogg (per the docs) and instead got 200 + a correct transcription, so I expanded the probe to .opus/.aac across all three STT models and both endpoints to confirm.

A follow-up commit will need to update the hand-written guide at mintlify-docs/docs/speech-to-text.mdx:189-193 (separate PR on the mintlify-docs repo) since that page lists the formats independently of the OpenAPI spec.

🤖 Generated with Claude Code

[github-actions]

✱ Stainless preview builds

This PR will update the togetherai SDKs with the following commit messages.

go

docs(api): update supported formats in audio transcription/translation file params

openapi

feat(api): add ogg/opus/aac format support to audio transcription and translation

python

docs(api): add .ogg, .opus, .aac to supported formats in audio transcriptions/translations

terraform

chore(internal): regenerate SDK with no functional changes

typescript

docs(api): update supported file formats in audio transcriptions/translations

✅ togetherai-openapi studio · code

Your SDK build had at least one "note" diagnostic.
generate ✅

⚠️ togetherai-go studio · code

Your SDK build had a failure in the test CI job, which is a regression from the base state.
generate ✅ → build ⏭️ → lint ✅ → test ❗

go get github.com/stainless-sdks/togetherai-go@3bf7862520e1abdc30136f58dc89dcff44ddad66

⚠️ togetherai-python studio · code

Your SDK build had at least one "warning" diagnostic.
generate ⚠️ → build ✅ → lint ✅ → test ⏭️

pip install https://pkg.stainless.com/s/togetherai-python/9c1211a8143dc435ef351dd57b9553d39ae06b8e/together-2.12.0-py3-none-any.whl

⚠️ togetherai-typescript studio · conflict

Your SDK build had at least one warning diagnostic.

✅ togetherai-terraform studio · code

Your SDK build had at least one "note" diagnostic.
generate ✅ → lint ✅ → test ✅

---

This comment is auto-generated by GitHub Actions and is automatically kept up to date as you push.
If you push custom code to the preview branch, re-run this workflow to update the comment.
Last updated: 2026-05-07 22:06:20 UTC