whisperbox 1.0.5

WhisperBox - Record and transcribe audio with ease

WhisperBox

A powerful command-line tool for transcribing and analyzing audio recordings with AI assistance. Record meetings, lectures, or any audio directly from your terminal and get instant transcriptions with summaries, sentiment analysis, and topic detection.

Available in two versions:

• Free: Command-line interface (CLI) version - Open source and MIT licensed
• Paid: GUI version with native desktop interface - One-time $10 purchase to support development

The GUI version offers the same powerful features in a user-friendly interface, perfect for those who prefer not to use the terminal. Purchase helps support ongoing development of free and open source AI tools. Purchase WhisperBox GUI version here

Features

• Live audio recording through terminal
• Multiple transcription models via Whisper AI
• AI-powered analysis including:
  • Meeting summaries and action items
  • Keynote presentation generation
  • Speech quality feedback
  • Modify voice input for any application
• Support for multiple AI providers:
  • Anthropic
  • OpenAI
  • Groq
  • Ollama (local models)
• Export to Markdown
• Rich terminal UI with color-coded output
• Configurable audio settings and output formats

Prerequisites

• Python 3.10 or higher
• FFmpeg (required for audio processing)
• Poetry (for dependency management)

Installation

1. Install FFmpeg if not already installed:

# On macOS using Homebrew
brew install ffmpeg

# On Ubuntu/Debian
sudo apt-get install ffmpeg

2. Install BlackHole for system audio capture (MacOS only)

brew install blackhole-2ch

3. Install portaudio for audio capture

brew install portaudio

4. Install whisperbox from pip

pip install whisperbox

Usage

Setup

The first time you run the app, you will go through the setup wizard.

wb

Then select the Whisper model you want to use. The smaller models are faster and quicker to download but the larger models are more accurate. Download times will vary depending on your internet speed.

Then select the AI provider you want to use. Ollama runs locally and does not require an API key.

Then select the model you want to use.

Then you will have the option to view the config file location so you can customize additional settings. This directory also contains the whisper models you downloaded and the data directory that contains all your recordings and transcriptions.

Basic Transcription

1. Start recording:

wb

2. Press Enter to stop recording when finished.

Advanced Options

• Specify a profile:

wb --profile monologue_to_keynote

• Specify a Whisper model:

wb --model large

• Enable full analysis (summary, sentiment, intent, topics):

wb --analyze

• Enable verbose output:

wb --verbose

Configuration

The config.yaml file allows you to customize:

• API settings for AI providers
• Audio recording parameters
• Transcription settings
• Output formats and directories
• Display preferences
• AI prompt templates

See the example config.yaml for all available options.

Extending WhisperBox

WhisperBox can be customized to handle your recordings exactly how you want. There are two main ways to extend it:

1. Profiles: Define what to do with your recordings
2. Scripts: Create custom actions for your profiles

Creating Custom Profiles

A profile is a simple YAML file that tells WhisperBox:

• What to do with your recording
• How to process the transcript
• Where to send the results

To create a profile:

1. Create a new .yaml file in the profiles/ folder (e.g., my_profile.yaml)
2. Add these three main sections:

# The name that appears in WhisperBox
name: my_profile

# Instructions for processing your recording
prompt: >
  Here's where you tell WhisperBox what to do with your recording.
  For example: "Create a summary with these key points..."

  The recording will appear here: {transcript}

# What to do with the results
actions:
  - script: output_to_markdown # Save as a file
  - script: copy_to_clipboard # Copy to clipboard

Built-in Actions

WhisperBox comes with several ready-to-use actions:

• output_to_markdown: Save as a Markdown file
• copy_to_clipboard: Copy to your clipboard
• output_to_terminal: Show in the terminal
• send_post_request: Send to a webhook URL

You can use multiple actions in a single profile:

actions:
  # Save the file
  - script: output_to_markdown
    config:
      filename: meeting_notes.md

  # Also copy it to clipboard
  - script: copy_to_clipboard

Creating Custom Scripts

Want to do something more custom? You can create your own action scripts:

1. Create a new Python file in the scripts/ folder (e.g., my_script.py)
2. Add a run_action function that handles the text:

def run_action(text, config):
    """
    text: The processed recording
    config: Any settings from your profile
    """
    print("I got the text:", text)
    # Do whatever you want with the text here!

3. Use it in your profile:

actions:
  - script: my_script
    config:
      any_setting: value

Example: Meeting Summary Profile

Here's a complete example that creates meeting summaries:

name: meeting_summary
prompt: >
  Create a clear summary of this meeting with:
  1. Key topics and decisions
  2. Action items and who's responsible
  3. Important deadlines or dates

  Meeting transcript: {transcript}

actions:
  # Save as a file
  - script: output_to_markdown
    config:
      filename: summary.md

  # Also copy to clipboard
  - script: copy_to_clipboard

Tips

• Test your profiles with short recordings first
• Check the profiles/ folder for more examples
• Use multiple actions to create powerful workflows
• Keep your prompts clear and specific

Contributing

1. Fork the repository
2. Create your feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add some amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

License

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

Acknowledgments

• Whisper AI for the transcription models
• Rich for the terminal UI
• All the AI providers supported by this tool

Authors

• Ty Fiero [e-mail]
• Mike Bird [e-mail]