@picovoice/porcupine-node

Picovoice Porcupine Node.js binding

Downloads in past

Stats

StarsIssuesVersionUpdatedCreatedSize
@picovoice/porcupine-node
2,76512.1.73 months ago2 years agoMinified + gzip package size for @picovoice/porcupine-node in KB

Readme

Porcupine Binding for Node.js

Porcupine

Porcupine is a highly accurate and lightweight wake word engine. It enables building always-listening voice-enabled applications using cutting edge voice AI.
Porcupine is:
  • private and offline
  • accurate
  • resource efficient (runs even on microcontrollers)
  • data efficient (wake words can be easily generated by simply typing them, without needing thousands of hours of bespoke audio training data and manual effort)
  • scalable to many simultaneous wake-words / always-on voice commands
  • cross-platform

To learn more about Porcupine, see the product, documentation, and GitHub pages.

Custom wake words

Porcupine includes several built-in keywords, which are stored as .ppn files. To train custom PPN files, see the Picovoice Console.
Unlike the built-in keywords, custom PPN files generated with the Picovoice Console carry restrictions including (but not limited to): training allowance, time limits, available platforms, and commercial usage.

Compatibility

This binding is for running Porcupine on Node.js 12+ on the following platforms:
  • Windows (x8664)
  • Linux (x8664)
  • macOS (x8664, arm64)
  • Raspberry Pi (2,3,4)
  • NVIDIA Jetson (Nano)
  • BeagleBone

Web Browsers

This npm package is for Node.js and does not work in a browser. Looking to run Porcupine in-browser? There are npm packages available for Web, and dedicated packages for Angular, React, and Vue.

AccessKey

Porcupine requires a valid Picovoice AccessKey at initialization. AccessKey acts as your credentials when using Porcupine SDKs. You can get your AccessKey for free. Make sure to keep your AccessKey secret. Signup or Login to Picovoice Console to get your AccessKey.

Usage

The binding provides the Porcupine class. Create instances of the Porcupine class to detect specific keywords.

Quick Start: Built-in keywords

The built-in keywords give a quick way to get started. Here we can specify that we want to listen for the wake words "grasshopper" and "bumblebee" with sensitivities of 0.5 and 0.65, respectively. Since Porcupine can listen to multiple keywords simultaneously, they are provided as an array argument.
const {
  Porcupine,
  BuiltinKeyword,
}= require("@picovoice/porcupine-node");

const accessKey = "${ACCESS_KEY}" // Obtained from the Picovoice Console (https://console.picovoice.ai/)

const handle = new Porcupine(
    accessKey,
    [BuiltinKeyword.GRASSHOPPER, BuiltinKeyword.BUMBLEBEE],
    [0.5, 0.65]);

// process a single frame of audio
// the keywordIndex provies the index of the keyword detected, or -1 if no keyword was detected
const keywordIndex = handle.process(frame);

List of built-in keywords

  • ALEXA
  • AMERICANO
  • BLUEBERRY
  • BUMBLEBEE
  • COMPUTER
  • GRAPEFRUIT
  • GRASSHOPPER
  • HEYGOOGLE
  • HEYSIRI
  • JARVIS
  • OKGOOGLE
  • PICOVOICE
  • PORCUPINE
  • TERMINATOR

Custom keywords

Providing an array of strings instead of the built-in enums allows you to specify an absolute path to a keyword .ppn file:
const accessKey = "${ACCESS_KEY}" // Obtained from the Picovoice Console (https://console.picovoice.ai/)

const handle = new Porcupine(
    accessKey,
    ["/absolute/path/to/your/keyword.ppn"],
    [0.5]);

Override model and library paths

The Porcupine constructor accepts two optional positional parameters for the absolute paths to the model and dynamic library, should you need to override them (typically, you will not).
const accessKey = "${ACCESS_KEY}" // Obtained from the Picovoice Console (https://console.picovoice.ai/)

const handle = new Porcupine(
  accessKey,
  keywordPaths,
  sensitivities,
  modelFilePath,
  libraryFilePath
);

Using the bindings from source

Unit Tests

Run yarn (ornpm install) from the binding/nodejs directory to install project dependencies. This will also run a script to copy all the necessary shared resources from the Porcupine repository into the package directory.
Run yarn test (or npm run test) from the binding/nodejs directory to execute the test suite.