STT CLI (Speech-to-Text Command Line Interface)

A command-line tool for real-time speech-to-text transcription with AI (Groq and OpenAI).

Features

• Real-time audio capture from microphone
• Support for multiple transcription providers:
  • Groq (using whisper-large-v3)
  • OpenAI (using Whisper)
• Efficient audio processing with proper chunking
• Clean shutdown handling with Ctrl+C

Installation

1. Make sure you have Rust installed on your system. If not, install it from rustup.rs

2. Clone the repository:

   git clone https://github.com/TwistingTwists/stt-cli
   cd stt-cli
   

3. Build the project:

   cargo build --release
   

Usage

The CLI supports different transcription providers through the -t or --transcription-provider flag:

# Using Groq
./target/release/stt-cli -t groq

# Using OpenAI
./target/release/stt-cli -t open-ai

Environment Variables

Before running the application, make sure to set up the required API keys:

• For Groq:

  export GROQ_API_KEY='your-groq-api-key'
  

• For OpenAI:

  export OPENAI_API_KEY='your-openai-api-key'
  

Expected Output

When running the application, you'll see:

1. Initialization messages for audio device setup
2. Real-time transcription of your speech
3. Status messages for audio processing and API requests

Example:

Initializing audio device...
Audio capture started. Speak into your microphone.
[Transcription] "Hello, this is a test of the speech to text system."
...

Press Ctrl+C to gracefully stop the application.

Contributing

Contributions are welcome! Please feel free to submit a Issue.

License

This project is licensed under either of

• Apache License, Version 2.0 (LICENSE-APACHE or http://www.apache.org/licenses/LICENSE-2.0)
• MIT license (LICENSE-MIT or http://opensource.org/licenses/MIT)

at your option.