Skip to content

refactor(media): extract STT from media-understanding into src/stt/#470

Merged
alexey-pelykh merged 1 commit intomainfrom
refactor/extract-stt-module
Mar 8, 2026
Merged

refactor(media): extract STT from media-understanding into src/stt/#470
alexey-pelykh merged 1 commit intomainfrom
refactor/extract-stt-module

Conversation

@alexey-pelykh
Copy link
Copy Markdown

Summary

  • Extract speech-to-text (STT) from src/media-understanding/ into a standalone src/stt/ module, mirroring the existing src/tts/ structure
  • Move audio transcription provider implementations (OpenAI, Deepgram, Google Gemini, Groq, Mistral), types, defaults, and provider query resolution into src/stt/
  • Media-understanding delegates audio provider execution to the new STT module while retaining shared infrastructure (attachment cache, CLI execution, scope gating)
  • Backward compatibility preserved via re-export shims in original media-understanding files

Closes #424

Test plan

  • All existing tests pass (10,423 tests across 1,211 test files)
  • New STT unit tests cover provider registry, transcription orchestration, provider query resolution
  • Telegram audio transcript test updated to mock from new import path
  • pnpm check passes (format + typecheck + lint)

🤖 Generated with Claude Code

…424)

Create a standalone speech-to-text module mirroring the existing src/tts/
structure. Audio transcription provider implementations (OpenAI, Deepgram,
Google Gemini, Groq, Mistral), types, defaults, and provider query resolution
are extracted into src/stt/. The media-understanding module delegates audio
provider execution to the new STT module while retaining shared infrastructure
(attachment cache, CLI execution, scope gating).

Backward compatibility is preserved via re-export shims in the original
media-understanding files. Consumer imports (Discord, Telegram) are updated
to import from the new location.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
@alexey-pelykh alexey-pelykh enabled auto-merge (squash) March 8, 2026 14:19
@alexey-pelykh alexey-pelykh merged commit 4579e9b into main Mar 8, 2026
5 checks passed
@alexey-pelykh alexey-pelykh deleted the refactor/extract-stt-module branch March 8, 2026 14:27
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

refactor(media): extract STT from media-understanding into src/stt/

1 participant