mindwork-transcribe

# Therapy Session Transcriber

Part of the **mindwork** suite. Converts therapy session recordings into clean, formatted transcripts.

## What It Does

1. **Chunks** large audio files at natural silence points (sentence boundaries)
2. **Transcribes** using OpenAI Whisper API
3. **Formats** as two-person conversation with **Me:** / **Therapist:** labels
4. **Corrects** grammar and transcription errors
5. **Translates** to English (for non-English sessions)

## Prerequisites

- Docker installed and running
- `OPENAI_API_KEY` environment variable set
- The `mindwork-transcribe` Docker image built (see Setup)

## Setup (One-Time)

Build the transcription Docker image from the plugin's transcribe directory:

```bash
# Find the mindwork plugin location and build the image
docker build -t mindwork-transcribe ~/src/mindwork/transcribe
```

Or if installed as a plugin, find the plugin path first:
```bash
# The transcribe tool is in the 'transcribe/' directory of this plugin
docker build -t mindwork-transcribe /path/to/mindwork/transcribe
```

## Usage

### Full Therapy Session Processing (Recommended)

Transcribe, format as conversation, and translate to English:

```bash
docker run --rm \
  -e OPENAI_API_KEY \
  -v $(pwd):/data \
  mindwork-transcribe /data/session.m4a --format-conversation --output /data/transcript.txt
```

### Raw Transcription Only

Just transcribe without formatting or translation:

```bash
docker run --rm \
  -e OPENAI_API_KEY \
  -v $(pwd):/data \
  mindwork-transcribe /data/session.m4a --output /data/transcript.txt
```

### With Speaker Diarization

For automatic speaker detection (alternative to --format-conversation):

```bash
docker run --rm \
  -e OPENAI_API_KEY \
  -v $(pwd):/data \
  mindwork-transcribe /data/session.m4a --diarize --output /data/transcript.txt
```

### Only Chunk (No Transcription)

Split a large file into chunks for later processing:

```bash
docker run --rm \
  -v $(pwd):/data \
  mindwork-transcribe /data/session.m4a --no-tr