YAML Metadata
Warning:
empty or missing yaml metadata in repo card
(https://huggingface.co/docs/hub/model-cards#model-card-metadata)
YouTube Video Transcriber
A Streamlit app that transcribes YouTube videos using Whisper, with optional formatting using a large language model, audio download, and video download.
How it works
- Downloads audio from YouTube videos using
yt-dlp. - Splits audio into speech segments using Silero VAD.
- Transcribes segments in batches using OpenAI's Whisper model.
- Formats the transcription using a large language model (if selected).
- Displays transcribed text with timestamps.
- Provides options to download the raw transcription, formatted transcription, audio, or video.
Requirements
Listed in requirements.txt
Usage
- Install dependencies:
pip install -r requirements.txt - Run the app:
streamlit run app.py - Enter a YouTube video URL.
- Choose options: Transcribe, Download Audio, Download Video, Format Text.
- Select a language or use auto-detect (under "Advanced Settings").
- Click "Process".
Screenshot
License
MIT "# YouTube-Transcriber"
Inference Providers
NEW
This model is not currently available via any of the supported Inference Providers.
The model cannot be deployed to the HF Inference API:
The model has no library tag.