Porcupine Wake Word
Python Quick Start
Platforms
- Linux (x86_64)
- macOS (x86_64, arm64)
- Windows (x86_64, arm64)
- Raspberry Pi (Zero, 3, 4, 5)
Requirements
- Picovoice Account & AccessKey
- Python 3.9+
- PIP
Picovoice Account & AccessKey
Signup or Login to Picovoice Console to get your AccessKey.
Make sure to keep your AccessKey secret.
Quick Start
Setup
Install Python 3.
Install the pvporcupine Python package using PIP:
Usage
Create an instance of Porcupine Wake Word that detects the included built-in wake words porcupine and bumblebee:
Pass in frames of audio to the .process function:
Release resources explicitly when done with Porcupine Wake Word:
Custom Keywords
Create custom keywords using the Picovoice Console.
Download the custom wake word file (.ppn) and create an instance of Porcupine Wake Word using the keyword_paths input argument:
Non-English Languages
Use the corresponding model file (.pv) to detect non-English wake words.
The model files for all supported languages are available
on the Porcupine Wake Word GitHub repository.
Pass in the model file using the model_path input argument to change the detection language:
Demo
For the Porcupine Wake Word Python SDK, we offer demo applications that demonstrate how to use the wake word engine on real-time audio streams (i.e. microphone input) and audio files.
Setup
Install the pvporcupinedemo Python package using PIP:
This package installs command-line utilities for the Porcupine Wake Word Python demos.
Usage
Use the --help flag to see the usage options for the demo:
Ensure you have a working microphone connected to your system and run the following command to detect the built-in keyword porcupine:
For more information on our Porcupine Wake Word demos for Python, head over to our GitHub repository.