https://github.com/roman01la/sub-deep

Transcribe and translate audio with AI
https://github.com/roman01la/sub-deep

deepl transcribe translate whisper

Last synced: about 2 months ago
JSON representation

Transcribe and translate audio with AI

• Host: GitHub
• URL: https://github.com/roman01la/sub-deep
• Owner: roman01la
• License: mit
• Created: 2023-04-24T18:54:43.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2023-04-24T18:55:38.000Z (over 1 year ago)
• Last Synced: 2024-05-02T04:40:55.743Z (8 months ago)
• Topics: deepl, transcribe, translate, whisper
• Language: JavaScript
• Homepage:
• Size: 6.84 KB
• Stars: 1
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
Transcribe audio and translate transcribed subtitles using [WhisperX](https://github.com/m-bain/whisperX) running on [JarvisLabs.ai](https://jarvislabs.ai/) for $0.49/hr and [DeepL](https://deepl.com) API. Takes 2-3 minutes of compute time to transcribe a short clip.

## Usage

1. Add your SSH key to your JarvisLabs.ai account at https://cloud.jarvislabs.ai/listsshkeys

2. Create `.env` file with the following content

```

DEEPL_KEY={your deepl.com API key}

JARVISLABS_KEY={your jarvislabs.ai API key}

JARVISLABS_UID={your jarvislabs.ai email address}

```

3. Install deps and run the script. This will spin up a small instance, upload audio file, transcribe it, download back and call DeepL API to translate subtitles.

```shell

yarn # install deps

# transcribe and translate

./transcribe.js audio.wav ./subs.srt uk en /path/to/id_rsa ssh-passphrase

```

CLI arguments in order:

1. WAV file

2. relative path to output file for SRT subtitles

3. Input language in the audio file

4. Output language for translated subtitles (will be saved on disk with `_{lang}` suffix in the filename)

5. Absolute path to your private SSH key

6. A pass phrase for your private key

You can retreive WAV audio from video file using `ffmpeg`

```shell

ffmpeg -i video.mp4 -vn -ac 1 audio.wav

```

_If you find this tool useful — consider endorcing my work with a small donation_

[![](https://www.buymeacoffee.com/assets/img/guidelines/download-assets-sm-1.svg)](https://www.buymeacoffee.com/romanliutikov)