https://github.com/imran-salim/transcriber
Transcribe speech using an LLM
https://github.com/imran-salim/transcriber
fastapi openai python
Last synced: 3 months ago
JSON representation
Transcribe speech using an LLM
- Host: GitHub
- URL: https://github.com/imran-salim/transcriber
- Owner: imran-salim
- License: mit
- Created: 2025-09-17T11:38:10.000Z (10 months ago)
- Default Branch: main
- Last Pushed: 2025-10-02T01:41:41.000Z (9 months ago)
- Last Synced: 2025-10-04T16:09:07.314Z (9 months ago)
- Topics: fastapi, openai, python
- Language: Python
- Homepage:
- Size: 32.2 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Transcriber
Real-time audio recording and transcription using FastAPI, PyAudio, and OpenAI.
## Setup
```bash
git clone https://github.com/imran-salim/transcriber.git
cd transcriber
python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
echo 'OPENAI_API_KEY=your_key' > .env
```
## Run
```bash
fastapi dev main.py
# http://localhost:8000/docs
```
## API
```
POST /record/start # Start recording
POST /record/stop/{id} # Stop & transcribe
GET /record/status/{id} # Check status
WS /record/ws # WebSocket interface
```
## Usage
```bash
# REST
curl -X POST http://localhost:8000/record/start
curl -X POST http://localhost:8000/record/stop/{session_id}
# WebSocket
websocat ws://localhost:8000/record/ws
> start
> stop
> quit
```
## Output
- Audio: `recording_{id}.wav`
- Text: `transcription_{id}.txt`
## License
MIT