vdt / buzz

Buzz transcribes audio from your computer's microphones to text using OpenAI's Whisper

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Buzz

Buzz

MIT License CI GitHub release (latest by date)

Buzz transcribes audio from your computer's microphones to text in real-time using OpenAI's Whisper.

Buzz - Watch Video

Requirements

To set up Buzz, first install ffmpeg (needed to run Whisper).

# on Ubuntu or Debian
sudo apt update && sudo apt install ffmpeg

# on MacOS using Homebrew (https://brew.sh/)
brew install ffmpeg

# on Windows using Chocolatey (https://chocolatey.org/)
choco install ffmpeg

# on Windows using Scoop (https://scoop.sh/)
scoop install ffmpeg

Installation

To install Buzz, download the latest version for your operating system. Buzz is available on Mac (Intel), Windows, and Linux.

Mac (Intel)

  • Download and open the *-mac.dmg file.
  • After the installation window opens, drag the Buzz icon into the folder to add Buzz to your Applications directory.

Windows

  • Download and extract the *-windows.tar.gz file.
  • Open the Buzz.exe file

Linux

  • Download and extract the *-unix.tar.gz file.
  • Open the Buzz binary file.

How to use

To record from a system microphone, select a model, language, task, microphone, and delay, then click Record.

Model: Default: Tiny.

Language: Default: English.

Task: Transcribe/Translate. Default: Transcribe.

Microphone: Default: System default microphone.

Delay: The length of time (in seconds) Buzz waits before transcribing a new batch of recorded audio. Increasing this value will make Buzz take longer to show new transcribed text. However, shorter delays cut the audio into smaller chunks which may reduce the accuracy of the transcription. Default: 10s.

For more information about the available model types, languages, and tasks, see the Whisper docs.

Record audio playing from computer

To record audio playing out from your computer, you'll need to install an audio loopback driver (a program that lets you create virtual audio devices). The rest of this guide will use BlackHole on Mac, but you can use other alternatives for your operating system (see LoopBeAudio, LoopBack, and Virtual Audio Cable).

  1. Install BlackHole via Homebrew

    brew install blackhole-2ch
  2. Open Audio MIDI Setup from Spotlight or from /Applications/Utilities/Audio Midi Setup.app.

    Open Audio MIDI Setup from Spotlight

  3. Click the '+' icon at the lower left corner and select 'Create Multi-Output Device'.

    Create multi-output device

  4. Add your default speaker and BlackHole to the multi-output device.

    Screenshot of multi-output device

  5. Select this multi-output device as your speaker (application or system-wide) to play audio into BlackHole.

  6. Open Buzz, select BlackHole as your microphone, and record as before to see transcriptions from the audio playing through BlackHole.

Build/run locally

To build/run Buzz locally from source, first install the dependencies:

  1. Install Poetry.

  2. Install the project dependencies.

    poetry install

Then, to run:

poetry run python main.py

To build:

poetry run pyinstaller --noconfirm Buzz.spec

About

Buzz transcribes audio from your computer's microphones to text using OpenAI's Whisper

License:MIT License


Languages

Language:Python 95.6%Language:Makefile 4.4%