Local macOS menubar voice assistant powered by MLX Whisper, Ollama, and Kokoro TTS
Project description
Ollama Vox
A local macOS menubar voice assistant that records speech, transcribes with MLX Whisper, gets responses from Ollama, and speaks back using Kokoro TTS.
Features
- Menubar app for easy access
- Local speech-to-text with
mlx-whisper - Local text generation with Ollama
- Local text-to-speech with
mlx-audio+ Kokoro voices - Status panel showing model info, rolling latency stats, and response history
Requirements
- macOS (Apple Silicon recommended)
- Python
3.12+ - Ollama installed and running locally
Installation
The easiest way to install is via pip or uv:
pip install ollama-vox
# or if using uv
uv tool install ollama-vox
If installing from source for development:
git clone https://github.com/maadhav-codes/ollama-vox.git
cd ollama-vox
uv sync
First-Time Setup
Download recommended STT + TTS models:
uv run ollama-vox --setup
Run
Start the menubar application:
uv run ollama-vox
Usage
- Click the microphone icon in your macOS menubar.
- Select Start Listening to speak.
- Select Stop Listening when you are done. The app will process your speech and respond with audio.
- Click Show Status to view latency, recent responses, and active models.
Configuration
Settings are managed in config.yaml.
audio: Adjust Voice Activity Detection (VAD) and recording limits.stt.model: Path to the local Whisper model.ollama: Set the endpoint, model name, and temperature.tts: Configure the voice, speaking rate, and Kokoro model path.
Community
We welcome contributions and feedback from the community!
- Contributing: Please see our Contributing Guide for local setup instructions and our PR review workflow.
- Reporting Issues: Encountered a bug or have a feature idea? Please use our Issue Forms.
- Discussions: Have questions or want to share how you're using Ollama Vox? Join the conversation in GitHub Discussions.
- Security: If you find a security vulnerability, please follow the steps in our Security Policy to report it privately.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file ollama_vox-1.0.3.tar.gz.
File metadata
- Download URL: ollama_vox-1.0.3.tar.gz
- Upload date:
- Size: 53.9 kB
- Tags: Source
- Uploaded using Trusted Publishing? Yes
- Uploaded via: twine/6.1.0 CPython/3.13.12
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
e0322f1b176978a4702c7215633edbb84f08ff65e93cc310d14c56705431d528
|
|
| MD5 |
8c052723ff7aada5b833b36fa97c0fe9
|
|
| BLAKE2b-256 |
c8f3929f5026a2aef1e14e400a58cb78256e6dc906da78c8e149853c6aa91991
|
Provenance
The following attestation bundles were made for ollama_vox-1.0.3.tar.gz:
Publisher:
ci_cd.yml on maadhav-codes/ollama-vox
-
Statement:
-
Statement type:
https://in-toto.io/Statement/v1 -
Predicate type:
https://docs.pypi.org/attestations/publish/v1 -
Subject name:
ollama_vox-1.0.3.tar.gz -
Subject digest:
e0322f1b176978a4702c7215633edbb84f08ff65e93cc310d14c56705431d528 - Sigstore transparency entry: 1451907360
- Sigstore integration time:
-
Permalink:
maadhav-codes/ollama-vox@44feefbc99af811fe1dbf052e615fce777d4b2b4 -
Branch / Tag:
refs/tags/v1.0.3 - Owner: https://github.com/maadhav-codes
-
Access:
public
-
Token Issuer:
https://token.actions.githubusercontent.com -
Runner Environment:
github-hosted -
Publication workflow:
ci_cd.yml@44feefbc99af811fe1dbf052e615fce777d4b2b4 -
Trigger Event:
release
-
Statement type:
File details
Details for the file ollama_vox-1.0.3-py3-none-any.whl.
File metadata
- Download URL: ollama_vox-1.0.3-py3-none-any.whl
- Upload date:
- Size: 60.8 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? Yes
- Uploaded via: twine/6.1.0 CPython/3.13.12
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
e49ac126c24e0af953a337781a007b1d460e507298b4a982ee1502231e4d83b7
|
|
| MD5 |
e77062c7a776e9bad92cee4412f2c632
|
|
| BLAKE2b-256 |
de85f8b88b3040c6d35006d00433e0064a2e0d252a109d0e2e4cc2f889b84780
|
Provenance
The following attestation bundles were made for ollama_vox-1.0.3-py3-none-any.whl:
Publisher:
ci_cd.yml on maadhav-codes/ollama-vox
-
Statement:
-
Statement type:
https://in-toto.io/Statement/v1 -
Predicate type:
https://docs.pypi.org/attestations/publish/v1 -
Subject name:
ollama_vox-1.0.3-py3-none-any.whl -
Subject digest:
e49ac126c24e0af953a337781a007b1d460e507298b4a982ee1502231e4d83b7 - Sigstore transparency entry: 1451907495
- Sigstore integration time:
-
Permalink:
maadhav-codes/ollama-vox@44feefbc99af811fe1dbf052e615fce777d4b2b4 -
Branch / Tag:
refs/tags/v1.0.3 - Owner: https://github.com/maadhav-codes
-
Access:
public
-
Token Issuer:
https://token.actions.githubusercontent.com -
Runner Environment:
github-hosted -
Publication workflow:
ci_cd.yml@44feefbc99af811fe1dbf052e615fce777d4b2b4 -
Trigger Event:
release
-
Statement type: