pediatric-ai-scribe-v3/docs/transcription-options.md
2026-05-08 19:08:19 +02:00

59 lines
1.8 KiB
Markdown

# Transcription Options
Ped-AI currently supports server-side transcription plus an explicit browser Web Speech preview option. Browser Whisper was removed and should not be offered in settings, documentation, public workers, or model download scripts.
## Recommended Clinical Setup
Use a server-side provider covered by your compliance requirements.
| Need | Recommended provider |
|---|---|
| HIPAA-eligible cloud STT | Google/Gemini through Vertex AI or AWS Transcribe with a BAA. |
| OpenAI-compatible routing | LiteLLM with a compliant upstream. |
| Direct OpenAI Whisper | Only when acceptable for your deployment. |
| Real-time draft preview | Browser Web Speech only with explicit user opt-in and privacy warning. |
## Configuration
```env
TRANSCRIBE_PROVIDER=litellm
LITELLM_API_BASE=https://your-litellm.example/v1
LITELLM_API_KEY=<key>
LITELLM_STT_MODEL=whisper-1
```
Other provider examples:
```env
# Google/Gemini
TRANSCRIBE_PROVIDER=google
GOOGLE_VERTEX_PROJECT=your-project-id
GOOGLE_STT_MODEL=gemini-2.0-flash
# AWS Transcribe
TRANSCRIBE_PROVIDER=aws
AWS_BEDROCK_REGION=us-east-1
AWS_TRANSCRIBE_MEDICAL=true
AWS_TRANSCRIBE_SPECIALTY=PRIMARYCARE
# Direct OpenAI Whisper
TRANSCRIBE_PROVIDER=openai
OPENAI_API_KEY=<key>
```
## Failure Handling
- Server transcription failures can create encrypted audio backups for retry.
- Users can retry or delete failed backups from Settings.
- Web Speech interim text is not a substitute for a server transcription response.
## Removed Paths
These should remain absent unless the project intentionally reintroduces browser-local STT with a new design review:
- `public/js/browserWhisper.js`
- `public/js/whisperWorker.js`
- `public/js/whisperWorkerV2.js`
- `public/models/Xenova/*`
- Browser Whisper setup/troubleshooting docs
- Whisper model download scripts for public browser models