Eagle Speaker Recognition
Web Quick Start

Platforms

• Chrome & Chromium-based browsers
• Edge
• Firefox
• Safari

Requirements

• Picovoice Account and AccessKey
• Node.js 14+
• npm

Picovoice Account & AccessKey

Signup or Login to Picovoice Console to get your AccessKey. Make sure to keep your AccessKey secret.

Overview

Eagle has two distinct steps: Enrollment and Recognition. In the enrollment step, Eagle analyzes a series of utterances from a particular speaker to learn their unique voiceprint. This step produces an Eagle Profile object, which can be stored and utilized during inference. During the Recognition step, Eagle compares the incoming frames of audio to the voiceprints of all enrolled speakers in real-time to determine the similarity between them.

Quick Start

Setup

1. Install Node.js .

2. Install the eagle-web package:

npm install @picovoice/eagle-web

Usage

Put the model file in the project's public directory or generate a base64 model using the build in script:

npx pvbase64 -i ${EAGLE_PARAMS_PATH} -o ${OUTPUT_FILE_PATH}

Speaker Enrollment

1. Create an instance of the EagleProfilerWorker:

import { EagleProfilerWorker } from "@picovoice/eagle-web";

const eagleModel = {
    publicPath: ${MODEL_RELATIVE_PATH},
    // or
    base64: ${MODEL_BASE64_STRING},
};

const eagleProfiler = await EagleProfilerWorker.create(
    "${ACCESS_KEY}",
    eagleModel);

2. Pass the enrollment audio data to the profiler until the enrollment percentage reaches 100%:

function getAudioData(numSamples): Int16Array {
    // get audio frame of size `numSamples`
}

let percentage = 0;
while (percentage < 100) {
    const audioData = getAudioData(eagleProfiler.minEnrollSamples);

    const result: EagleProfilerEnrollResult = await eagleProfiler.enroll(audioData);
    console.log(EagleProfilerEnrollFeedback[result.feedback]);
    percentage = result.percentage;
}

3. Export the speaker profile:

const speakerProfile: EagleProfile = eagleProfiler.export();

4. Release the resources acquired by the profiler:

eagleProfiler.release();

Speaker Recognition

1. Create an instance of EagleWorker with one or more speaker profiles:

import { EagleWorker } from "@picovoice/eagle-web";

const eagleModel = {
    publicPath: ${MODEL_RELATIVE_PATH},
    // or
    base64: ${MODEL_BASE64_STRING},
};

const eagle = await EagleWorker.create(
    "${ACCESS_KEY}",
    eagleModel,
    speakerProfile);

2. Pass an incoming audio stream to Eagle:

function getAudioData(numSamples): Int16Array {
  // get audio frame of size `numSamples`
}

while (true) {
  const audioData = getAudioData(eagle.frameLength);
  const scores: number[] = await eagle.process(audioData);
}

3. Release the resources acquired by Eagle:

eagle.release();

Demos

For the Eagle Web SDK, we offer a demo application that demonstrate how to use the speaker recognition engine on a real-time audio stream.

Setup

Clone the Eagle repository from GitHub:

git clone --recurse-submodules https://github.com/Picovoice/eagle.git

Usage

1. Install dependencies and run:

cd eagle/demo/web
npm install
npm run start

2. Open http://localhost:5000 to view it in the browser.

Resources