Skip to content

Latest commit

 

History

History
34 lines (25 loc) · 1.21 KB

README.md

File metadata and controls

34 lines (25 loc) · 1.21 KB

Local Voice Chat MVP with Tool-use and Memory

A voice-based chat application that allows users to interact with an AI assistant using speech. The application leverages OpenAI's Whisper small for accurate speech recognition and Kokoro-TTS for natural-sounding voice synthesis. You can use a LM Studio local model that needs to be served at http://localhost:1234 to generate response. Requires meta-llama-3.1-8b-instruct for tool-use. Works great on Macbook M1 Pro, but Linux works too.

Use python3.12 to run this application, because of PyTorch dependency.

Hacked together with Claude and Cursor.

Quick Start

  1. Clone the repository:

    $ git clone https://github.com/jpzk/voicemvp.git
$ cd voicemvp

  2. Install system dependencies (macOS):

    $ brew install portaudio

  3. Install dependencies:

    $ python3.12 -m venv env
$ source env/bin/activate
$ python3.12 -m pip install -r requirements.txt

  4. Run the application:

    $ python3.12 voice_chat_agent.py

  5. The first run might take a while as it needs to download the models.