Advanced Speech Transcription App

Table of Contents

• Introduction 📖
• Key Features 💫
• Getting Started 🌱
  • Prerequisites 📋
  • Installation 📥
• How to Use 💻
• Customization 🔧
• Contributing 🤝
• License 📄
• Support and Contact 🙌

Introduction 📖

The Advanced Speech Transcription App aims for offline, precise yet fast speech recognition using multiple threads for each phrase recorded.

It uses Whisper models locally.

Key Features 💫

• State-of-the-Art Transcription: Utilizes OpenAI's Whisper models to deliver exceptionally accurate transcriptions.
• Real-Time Recognition: Immediate transcription with real-time voice processing capabilities.
• Adaptive Noise Reduction: Reduces background noise and enhances speech clarity.
• Multi-Threaded Background Processing: Splits speech recognition into as many threads as needed depending on the speaker's type of speech.
• Custom Stop Commands: Easily terminate transcription sessions with user-defined phrases.
• Persistent Output: Automatically saves the transcribed text to your chosen destination.

Getting Started 🌱

Prerequisites 📋

Before installing the Advanced Speech Transcription App, you'll need:

• Python 3.9 or above
• Pip package manager (usually included with Python)
• Microphone (duh 😂)

Installation 📥

# Clone the project repository
git clone https://github.com/AlexYelisieiev/advanced-speech-transcription.git

# Navigate to the project directory
cd advanced-speech-transcription-app

# Create and activate a virtual environment
python3 -m venv venv
venv/bin/activate

# Install all dependencies
pip install -r requirements.txt

# Launch the application
python main.py

Note

Once launched for the first time, it will install the Whisper model to use locally.

How to Use 💻

Initiate the app and expose it to the speech source once it says "Listening:" (until then, it's adjusting the microphone to the environment). The real-time transcription will appear on your screen. To end the transcription, simply utter a predefined stop phrase, and the session will conclude, saving the transcription to the designated file.

Customization 🔧

Personalize your experience by editing the SpeechRecognizer class parameters:

• whisper_model: Select from various Whisper model options to match your transcription accuracy and language needs.
• output_file_path: Designate your desired file path for storing transcripts.
• STOP_PHRASES: Customize the stop phrases for an intuitive way to end transcription sessions.

Contributing 🤝

New PRs, feedback, etc. are always welcome. Also, if you want to chat or something, DMs are open 😉

License 📄

Distributed under the MIT License. See LICENSE for more information.

Support and Contact 🙌

For support, feature requests, or to report a bug, please submit an issue on the GitHub repository. Create PRs if you want to add something yourself 🤝

Made with 💙 by Alex Yelisieiev