AI Transcriber is a Python-based tool that converts speech to text using OpenAI’s Whisper model. Designed for content creators, podcasters, and musicians, this tool ensures high-accuracy captions, even for fast speech, low-quality audio, or background noise.
✅ AI-Powered Speech-to-Text - Uses Whisper for precise transcription.
✅ Handles Noisy Audio & Fast Speech - Great for rap, interviews, or podcasts.
✅ Editable Captions - Outputs a text file for easy review and editing.
✅ Simple Command-Line Interface (CLI) - Just run the script and get your transcript.
✅ Future Plans - AI-assisted accuracy review, web UI, YouTube plugin.
- Clone the repository:
git clone https://github.com/your-username/ai-transcriber.git cd ai-transcriber - Install dependencies:
pip install openai-whisper torch soundfile
- Run the Script:
python transcribe.py your_audio_file.mp3
- Place your audio file (mp3, wav, or m4a) in the project folder.:
- Run:
python transcribe.py example.mp3
- The transcription will appear in the terminal and be saved as example.mp3.txt:
🔹 AI-powered accuracy review to catch mistakes.
🔹 Web App & Browser Extension for YouTube captions.
🔹 Custom Model Fine-Tuning for better slang/rap/music transcription.
Want to improve this? Fork the repo, submit PRs, or suggest ideas in the Issues tab!
MIT License - Free to use and modify.