An example project using local voice activity detection and Carter in Python. This will run on a Raspberry Pi, Macbook, Windows PC and gives you complete voice input and output to your character. It also uses the Carter API to get the agent's voice output (subject to improvement soon!)
B13527BD-CC65-486E-97F0-535DCDDA6688.mp4
First, install the other requirements:
pip install -r requirements.txt
Then, run the following command to start the server:
python app.py --key your-api-key --user=UNIQUE_STRING
To find out more about the API key and configure your agent, visit the Carter website.
to install PyAudo for M1 Macs, this will require a small workaround, as PortAudio is not automatically detected. first, we need to install PortAudiobrew install portaudio
then, we need to create a new file:
nano ~/.pydistutils.cfg
the contents of that file may vary slightly (ie, version number)
[build_ext]
include_dirs=/opt/homebrew/Cellar/portaudio/19.7.0/include/
library_dirs=/opt/homebrew/Cellar/portaudio/19.7.0/lib/
where 19.7.0 should be replaced with the portaudio version you get installed.
this should be executed BEFORE running the pip install -r requirements.txt
command