TranscribeIt is a free software transcription service for online videos, live streams and local multimedia content.
Hard of hearing and speech population, people with visual impairments and sensory processing disorders have difficulty navigating video content. With online platforms providing little to no transcriptions or inaccurate transcriptions, especially for live streams, it hinders acccessibility. TransribeIt aims to provide customized and diarized transcriptions for making video content accessible.
- Transcribe videos and livestreams from YouTube and other online sources supported by yt-dlp.
- Get transcriptions with timestamps in an intuitive UI for multiple languages using faster-whisper
- Download transcripts in different formats (STT, JSON, etc.)
- Include speaker diarization
- Background processing
- Transcription from local audio file uploads
- Clone the project
git clone https://github.com/fossiaorg/transcribeit
cd transcribeit- Set up frontend
cd web
yarn install- Configure environment variables for frontend
cp .env.sample .env
# Edit the values as per needed for .env- Set up backend
cd server
uv sync- Configure environment variables for backend
cp .env.sample .env
# Edit the values as per needed for .env- Run the frontend and backend
- For frontend
yarn dev
- For backend
source .venv/bin/activate fastapi dev src/transcribeit/server.py
The frontend should be running at: http://localhost:3000.
The backend should be running at: http://localhost:8000.
You can spread the word about the project on social media, show your love by starring the repository, help with design and documentation, or contribute code.
For more information on contributing code, check out our contributing guide.
TranscribeIt would not exist without these amazing projects:
- yt-dlp for supporting multiple online video streaming platforms
- faster-whisper for timestamped, mutli-lingual transcriptions
- Chakra UI for accessible UI components
- FastAPI for rapid API development
- Next.js for frontend framework
TranscribeIt is licensed under GNU Affero General Public License version 3. For more information, check out our LICENSE file.