git clone --recursive [email protected]:davesarmoury/PotatOS.git
cd PotatOS
pip3 install -r requirements.txt
sudo groupadd -f -r gpio
sudo usermod -a -G $USER
sudo cp lib/python/Jetson/GPIO/99-gpio.rules /etc/udev/rules.d/
#sudo reboot
bash jetson-containers/install.sh
cd patch/
./patch_platformdetect.sh
cd ..
jetson-containers run --name ollama $(autotag ollama)
ollama run llama3.2:3b --keepalive 60m
cd wiki_rag/
pip3 install -r requirements.txt
#python3 wiki_llamaindex_preprocess.py
#python3 wiki_llamaindex.py
python3 wiki_ollama.py
See https://developer.nvidia.com/embedded/learn/tutorials/connecting-bluetooth-audio, then pair a bluetooth speaker with bluetoothctl. Then set the volume
sudo nano /lib/systemd/system/bluetooth.service.d/nv-bluetooth-service.conf
amixer -D pulse sset Master 50%
pip3 install -r vosk-server/requirements.txt
pip3 install sounddevice
sudo apt install libportaudio2
cd vosk-server/websocket
wget https://alphacephei.com/vosk/models/vosk-model-small-en-us-0.15.zip
unzip vosk-model-small-en-us-0.15
rm vosk-model-small-en-us-0.15.zip
mv vosk-model-small-en-us-0.15 model
python3 asr_server.py
Train or download the GLaDOS Piper voice. See train_piper for more details. If you don't want to train you own, you can download pre-trained models using
cd jetson-containers/data/models/piper/
wget https://huggingface.co/DavesArmoury/GLaDOS_TTS/resolve/main/glados_piper_medium.onnx.json
wget https://huggingface.co/DavesArmoury/GLaDOS_TTS/resolve/main/glados_piper_medium.onnx
On the jetson, put the onnx and onnx.json files into jetson-containers/data/models/piper, then run the command below
docker pull dustynv/piper-tts:r36.2.0
jetson-containers run $(autotag piper-tts) python3 -m piper.http_server --port 5001 -m /data/models/piper/glados_piper_medium.onnx
Run a quick test with
curl -G --data-urlencode 'text=I like big butts, I cannot lie.' --output - 'localhost:5001' | aplay
python3 potatos.py -r 44100 -d 0
PulseAudio will automatically disconnect when no one is logged into the machine. That means if you are trying to run it headless, your audio will break. This is gross, but just ssh the terminal into itself so the machine sees a "remote" connection
ssh localhost