https://github.com/orhancavus/transcribe_video

Extract Subtitles from YouTube Videos with OpenAI Whisper and Insanely Fast Whisper
https://github.com/orhancavus/transcribe_video

insanely-fast speach-to-text whisper

Last synced: about 16 hours ago
JSON representation

Extract Subtitles from YouTube Videos with OpenAI Whisper and Insanely Fast Whisper

• Host: GitHub
• URL: https://github.com/orhancavus/transcribe_video
• Owner: orhancavus
• Created: 2025-01-08T22:43:10.000Z (about 21 hours ago)
• Default Branch: main
• Last Pushed: 2025-01-08T23:01:12.000Z (about 21 hours ago)
• Last Synced: 2025-01-09T00:21:49.291Z (about 20 hours ago)
• Topics: insanely-fast, speach-to-text, whisper
• Language: Python
• Homepage:
• Size: 20.4 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Extract Subtitles from YouTube Videos with OpenAI Whisper and Insanely Fast Whisper

Author : Orhan Cavus

Date   : January 2025

## Download Audio from YouTube

To download audio from a YouTube video, use the following command:

```bash

yt-dlp -x --audio-format mp3 -o "input/video_audio.%(ext)s" 

```

Example:

```bash

yt-dlp -x --audio-format mp3 -o "input/video_audio.%(ext)s" https://www.youtube.com/watch?v=0Vjh5d5rez0

```

## Extract Text from Audio with Whisper

Use the following commands to extract text from the downloaded audio using Whisper:

```bash

whisper input/video_audio.mp3 --model medium

whisper input/video_audio.mp3 --model large --language English --output_format srt

whisper input/video_audio.mp3 --model medium --output_format -f {all} --output_dir output

whisper input/video_audio.mp3 --model medium -f all --output_dir output

whisper input/video_audio.mp3 --model medium --task translate -f srt --output_dir output

```

## Extract Text from Audio with Insanely Fast Whisper

For faster transcription, use Insanely Fast Whisper:

[Insanely Fast Whisper GitHub Repository](https://github.com/Vaibhavs10/insanely-fast-whisper)

```bash

insanely-fast-whisper --file-name input/video_audio.mp3 --transcript-path output/output_new.srt --device mps

```

## Installation

Install the required packages with the following commands:

```bash

pip install openai-whisper

pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/nightly/cpu

```

## Additional Links

- [YouTube to Transcript](https://youtubetotranscript.com/)