Installation

1. Clone the repository:

   git clone https://github.com/rick12000/vocalance.git
   cd vocalance
   

2. Create a Python 3.13.9 environment and activate it.

3. Install (choose one):

   Base install — no LLM / smart-dictation features:

   pip install .
   

   With LLM features — smart dictation and text-amend (requires Microsoft C++ Build Tools with “Desktop development with C++” workload):

   pip install ".[llm]"
   

   When using UV with the lock file instead of pip:

   uv sync               # base only
   uv sync --extra llm   # base + LLM features
   

4. Run:

   python vocalance.py
   

On first launch, Vocalance downloads required assets: the YAMNet sound model for custom sound recognition and, if LLM features are enabled, the LLM bundle used for smart dictation. Allow several minutes for these downloads. The Vosk and Moonshine speech models are bundled and require no separate download.

Note

LLM features (smart dictation, text-amend) are optional. When not installed, the smart-dictation UI sections and their voice triggers are automatically hidden. All other features work without any LLM dependency.