hobokenchicken 0e74a16b40 fix(stt): revert to reliable REST gpt-4o-transcribe + MediaRecorder full-blob (Realtime WS not accessible on key) ...

- Backend: added transcribe_audio (gpt-4o-transcribe), switched audio handler to full blob -> REST -> LLM -> streaming TTS
- Frontend: MediaRecorder (webm/opus) full recording sent on stop (one blob per utterance)
- Removed dead WhisperStream callbacks and pending_transcript/lock
- This unblocks voice per AUDIT item 1 (Option B fallback). Deltas will come in later item.
- Also preps for deprecation fix (MediaRecorder is the good path).

2026-06-04 15:23:57 -04:00

..

models

init: Kira — AI body double with Honcho memory

2026-06-04 10:51:38 -04:00

routers

init: Kira — AI body double with Honcho memory

2026-06-04 10:51:38 -04:00

services

fix(stt): try gpt-4o-realtime-preview as base session model + gpt-realtime-whisper for input_audio_transcription (per OpenAI error guidance)

2026-06-04 15:16:08 -04:00

__init__.py

init: Kira — AI body double with Honcho memory

2026-06-04 10:51:38 -04:00

config.py

init: Kira — AI body double with Honcho memory

2026-06-04 10:51:38 -04:00

Dockerfile

init: Kira — AI body double with Honcho memory

2026-06-04 10:51:38 -04:00

main.py

fix(stt): revert to reliable REST gpt-4o-transcribe + MediaRecorder full-blob (Realtime WS not accessible on key)

2026-06-04 15:23:57 -04:00

requirements.txt

feat: OpenAI Realtime API pipeline

2026-06-04 13:32:39 -04:00