https://github.com/otonomee/mic2transcript

CLI tool that continuously transcribes audio from the device's built-in microphone to a text file. Runs in the background, providing an ongoing log of ambient audio as text.
https://github.com/otonomee/mic2transcript

audio cli cli-tool openai speech speech-transcription transcription whisper

Last synced: 3 months ago
JSON representation

CLI tool that continuously transcribes audio from the device's built-in microphone to a text file. Runs in the background, providing an ongoing log of ambient audio as text.

• Host: GitHub
• URL: https://github.com/otonomee/mic2transcript
• Owner: otonomee
• Created: 2024-06-21T21:44:04.000Z (7 months ago)
• Default Branch: main
• Last Pushed: 2024-06-22T05:19:06.000Z (7 months ago)
• Last Synced: 2024-09-27T06:22:15.342Z (4 months ago)
• Topics: audio, cli, cli-tool, openai, speech, speech-transcription, transcription, whisper
• Language: Python
• Homepage:
• Size: 15.6 KB
• Stars: 2
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# mic2transcript

mic2transcript is a CLI tool that will continuously transcribe audio from the device's built-in microphone to a text file using [OpenAI Whisper](https://github.com/openai/whisper).

## CLI Inputs 

- Name of output text file

- Transcription model

    - `tiny`: Fastest, lowest accuracy. Suitable for quick transcriptions or resource-constrained environments.

    - `medium`: Balanced performance. Good for general use, offering a trade-off between speed and accuracy.

    - `large`: Highest accuracy, slowest speed. Best for scenarios requiring maximum transcription quality.

## Requirements

- Python 3.6+

- sounddevice

- numpy

- whisper

## Installation

1. Clone this repository: `https://github.com/otonomee/continuous-mic-transcribe`

2. Install the required packages: 

```

pip install -r requirements.txt

```

## Usage

To start the transcription, run the tool from the terminal: 

```

python main.py

```

You will be prompted to select a Whisper model and specify an output file name for the transcriptions.

The tool will run in the background, transcribing any detected audio into the specified text file. Press 'q' in the terminal to stop 

the transcription process.