Timestamper

Timestamper is a Python tool for working with audio transcription and subtitle files. It enables you to generate timestamped templates for text, transcribe audio into subtitles with word-level timestamps, and convert .srt subtitle files into .docx documents.

Features

• Add Timestamps to Text Files: Automatically generate .srt-style timestamp templates for .txt files.
• Audio Transcription: Transcribe audio files (.mp3, .wav, .m4a, .aac, .webm) into .srt subtitle files using the faster-whisper library.
• Word-Level Timestamps: Generate subtitles with precise word-level timestamps.
• SRT to DOCX Conversion: Convert .srt subtitle files into .docx documents with timestamps and text.
• Custom Whisper Model Selection: Choose from various Whisper model sizes (tiny, base, small, medium, large) for transcription.
• Multi-Language Support: Specify the language for audio transcription.

Installation

1. Clone the repository:

   git clone https://github.com/leopalladium/timestamper.git
   cd timestamper

2. Install dependencies:

   pip install -r requirements.txt

Usage

Transcribing Audio Files

1. Run main.py.
2. Select the root directory containing your audio files.
3. Choose the Whisper model size.
4. Enter the language code (e.g., en, ru).
5. Select which audio files to process or type all to process all found files.
6. The script generates .srt files with word-level timestamps in the same directory as the audio files.

Adding Timestamp Templates to Text Files

1. Place your .txt file in the script directory.
2. Use the add_timestamps_to_sentences function to generate a timestamped template.

Converting SRT to DOCX

1. Use the convert_srt_to_docx function to convert an .srt file to a .docx document.

Roadmap

• Add speaker diarization
• Optimize and debug code
• Develop as a subtitle editing tool
• API for server deployment
• Add audio track from video text recognition
• Make executable more lightweight
• Add signature for executable

License

This project is licensed under the MIT License. See the LICENSE file for details.

Author

• Klimentsi Katsko (@leopalladium)