rodrigokrosa / personal-assistant

Steps when initiating a new ML python project

1. Create folder structure

2. Configure pyenv

pyenv local 3.10

3. Initialize poetry

poetry init

4. Initialize pre-commit

pre-commit install

TTS HTTP Server with Piper TTS

1. Clone Piper-TTS repository

git clone https://github.com/rhasspy/piper.git

2. Go to python_run folder

cd piper/src/python_run

3. Install piper-tts

pip install -e .

4. Install requirements for the http_server

pip install -r requirements_http.txt

5. Download onxx pretrained models from the repository [link]

6. Run http_server with:

screen python -m piper.http_server --model <model_name>

LLM HTTP Server with Ollama

1. Install ollama

curl -fsSL https://ollama.com/install.sh | sh

2. Init ollama server

screen -S llm ollama serve

3. Test and download llm model

ollama run llama3