The sh.edict/edict MCP server provides search over dictation transcripts and transcription of audio or video. It performs transcription on-device and outputs speaker labels, enabling users to locate relevant transcript content and produce labeled transcriptions without relying on external services.
🛠️ Key Features
- Search dictation transcripts
- Transcribe audio or video
- On-device transcription
- Speaker labels in transcription output
🚀 Use Cases
- Finding specific sections within dictation transcripts
- Creating transcripts from recorded audio/video
- Producing speaker-attributed transcripts for review
⚡ Developer Benefits
- Consistent transcript content via searchable dictation outputs
- Labeled transcription suitable for downstream processing that needs speaker attribution