easytranscribe 0.1.2

Easy speech-to-text transcription from audio files or live microphone input using Whisper.

EasyTranscribe

A simple Python-based voice assistant that captures speech from your microphone or from recorded file, detects silence, and transcribes spoken words to text using OpenAI Whisper. Easily extensible for integration with LLMs like Ollama or Gemma.

Features

• Real-time microphone audio capture
• Automatic silence detection and recording stop
• Speech-to-text transcription using Whisper
• Comprehensive transcription logging with detailed metrics
• Easy integration with other AI models

Installation

1. Clone the repository:

   git clone https://github.com/akhshyganesh/easytranscribe.git
   cd easytranscribe
   

2. Install dependencies:

   pip install -r requirements.txt
   

Usage

Run the main script:

python main.py

Speak into your microphone. The assistant will automatically stop recording after a few seconds of silence and transcribe your speech.

easytranscribe

Easy speech-to-text transcription from audio files or live microphone input using OpenAI's Whisper.

✨ Features

• 🎤 Live microphone transcription with automatic silence detection
• 📁 Audio file transcription supporting multiple formats
• 📊 Automatic logging with timestamps and performance metrics
• 🔧 Simple CLI interface for quick usage
• 🐍 Easy Python API for integration into your projects
• 📈 Log analysis tools to view transcription history and statistics

🚀 Quick Start

Installation

pip install easytranscribe

Python API

Live microphone transcription:

from easytranscribe import capture_and_transcribe

# Start live transcription (speaks and waits for silence)
text = capture_and_transcribe(model_name="base")
print(f"You said: {text}")

Audio file transcription:

from easytranscribe import transcribe_audio_file

# Transcribe an audio file
text = transcribe_audio_file("path/to/audio.wav", model_name="base")
print(f"Transcription: {text}")

View transcription logs:

from easytranscribe import view_logs

# View today's logs with statistics
logs = view_logs(date="today", stats=True)
print(f"Total entries: {logs['total_count']}")

Command Line Interface

Live transcription:

easytranscribe live --model base

File transcription:

easytranscribe file path/to/audio.wav --model base

View logs:

# View today's logs
easytranscribe logs --date today --stats

# View last 10 entries
easytranscribe logs --tail 10

# List available log dates
easytranscribe logs --list-dates

📋 Available Whisper Models

Model	Size	Speed	Accuracy	Use Case
tiny	39MB	Fastest	Good	Real-time, low resource
base	74MB	Fast	Better	Balanced performance
small	244MB	Medium	Good	Higher accuracy
medium	769MB	Slow	Very Good	Professional use
large	1550MB	Slowest	Best	Maximum accuracy
turbo	809MB	Fast	Excellent	Best balance (default)

🔧 Configuration

Audio Settings (Live Recording)

The package automatically handles:

• ✅ Silence detection (3 seconds of silence stops recording)
• ✅ Minimum recording time (2 seconds)
• ✅ Audio level monitoring
• ✅ Automatic microphone input

Logging

Transcriptions are automatically logged to logs/transcription_YYYY-MM-DD.log with:

• 📅 Timestamp
• 🤖 Model used
• ⏱️ Processing time
• 🎵 Audio duration (for live recording)
• 📝 Transcribed text

🛠️ Development

Install from Source

git clone https://github.com/akhshyganesh/easytranscribe.git
cd easytranscribe
pip install -e .

Run Tests

python test/test_integration.py

📄 Requirements

• Python 3.8+
• OpenAI Whisper
• sounddevice (for microphone input)
• numpy

📖 Documentation

For comprehensive documentation, examples, and API reference, visit:

🌐 EasyTranscribe Documentation

The documentation includes:

• 🚀 Quick Start Guide
• 💻 CLI Usage
• 🐍 Python API
• 📝 Examples
• ⚙️ Configuration
• 🔧 Advanced Usage

🤝 Contributing

Contributions are welcome! Please see CONTRIBUTING.md for guidelines.

📜 License

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

🙏 Acknowledgments

• OpenAI Whisper for the amazing speech recognition model
• sounddevice for microphone input handling