WhisperChain - an AI-powered real-time speech-to-text tool that automatically cleans and optimizes text.

What is WhisperChain?

WhisperChain is an open-source speech recognition tool that improves work efficiency based on voice input. WhisperChain uses Whisper.cpp for real-time speech recognition, converting speech to text. It then uses LangChain to clean and optimize the text, removing filler words and refining the content. WhisperChain supports global keyboard shortcuts, allowing users to easily initiate voice input at any time and automatically copying the cleaned text to the clipboard for convenient pasting. WhisperChain provides a Streamlit Web UI and FastAPI server architecture, supporting various configurations and extensions, making it suitable for users who require efficient voice recording and text optimization.

WhisperChain's main functions

• Real-time speech recognitionReal-time speech-to-text conversion.
• Text cleaning and optimizationThe process involves processing the identified text to remove filler words (such as "um" and "ah"), optimizing the syntax, and refining the content.
• Global hotkey supportCustomizable global hotkeys (such as <Ctrl>+<Alt>+R) allow for quick and easy activation of voice input.
• Automatic clipboard integrationThe cleaned text is automatically copied to the clipboard, allowing users to paste it wherever needed.
• Streamlit Web UIIt provides a graphical interface, allowing users to interact with tools via a webpage, view history, or re-edit text.
• Flexible configuration optionsIt supports custom settings based on configuration files or command-line parameters, such as selecting different speech recognition models and ports.

WhisperChain's technical principles

• Whisper.cppWhisper.cpp is an open-source speech recognition engine based on OpenAI's Whisper model. It converts speech signals into text, supports multiple languages and dialects, and provides efficient real-time speech recognition capabilities. It is one of the core components of WhisperChain.
• LangChainLangChain is a tool for text processing and optimization. It cleans the recognized text, removing unnecessary filler words and duplicate content, and optimizing grammar and expression. It also performs semantic optimization based on context, resulting in more fluent and natural-sounding final text.
• FastAPI Server and WebSocketWhisperChain uses FastAPI to build the backend server and WebSocket for real-time communication. Voice data is transmitted to the server via WebSocket, where Whisper.cpp performs speech recognition and LangChain processes the text. The processed text is then returned to the client, achieving efficient data processing and transmission.
• Streamlit Web UIStreamlit is a framework for rapidly developing web applications. WhisperChain uses Streamlit to build a graphical interface, allowing users to interact with tools via web pages, easily view history, re-edit text, or adjust settings.
• Global hotkeys integrated with clipboardBased on system-level global hotkey monitoring, users can initiate voice input at any time. The processed text is automatically copied to the system clipboard for easy pasting and use in other applications.

WhisperChain project address

• GitHub repository:https://github.com/chrischoy/WhisperChain

Application scenarios of WhisperChain

• Meeting minutesQuickly convert meeting audio into concise text for easy summarization of key points.
• Writing aidsTransform spoken ideas into written text to improve writing efficiency.
• Remote collaborationRecord meeting content in real time and share it quickly to improve communication efficiency.
• Content creationTranslate spoken content into scripts, optimize language expression, and assist in video or podcast production.
• Personal NotesRecord your thoughts anytime, anywhere with your voice, automatically convert them to text, and quickly paste them.