Leopard Speech-to-Text
Web Quick Start

Platforms

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

Requirements

• Picovoice Account and AccessKey
• Node.js 16+
• npm

Picovoice Account & AccessKey

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

Quick Start

Setup

1. Install Node.js.

2. Install the Leopard Speech-to-Text Web package:

npm install @picovoice/leopard-web

Usage

Generate a custom Leopard Speech-to-Text model from Picovoice Console or download a default model.

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

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

Create a LeopardWorker instance using a base64 model or a model hosted in a public directory:

import { LeopardWorker } from "@picovoice/leopard-web"
import leopardParams from "${LEOPARD_PARAMS_BASE64_PATH}";

const leopard = await LeopardWorker.create(
  "${ACCESS_KEY}",
  {
    base64: leopardParams,
    // or
    publicPath: "${MODEL_RELATIVE_PATH}"
  }
);

Transcribe audio (sample rate of 16 kHz, 16-bit linearly encoded and 1 channel):

function getAudioData(): Int16Array {
  // get audio
  return new Int16Array();
}

const { transcript, words } = await leopard.process(getAudioData());

Release resources explicitly when done with Leopard:

await leopard.release()

Non-English Languages

In order to use Leopard with other languages, you need to use the corresponding model file (.pv) for the desired language. The model files for all supported languages are available on the Leopard GitHub repository.

Word Metadata

Along with the transcript, Leopard Speech-to-Text returns metadata for each transcribed word. Available metadata items are:

• Start Time: Indicates when the word started in the transcribed audio. Value is in seconds.
• End Time: Indicates when the word ended in the transcribed audio. Value is in seconds.
• Confidence: Leopard Speech-to-Text's confidence that the transcribed word is accurate. It is a number within [0, 1].
• Speaker Tag: If speaker diarization is enabled on initialization, the speaker tag is a non-negative integer identifying unique speakers, with 0 reserved for unknown speakers. If speaker diarization is not enabled, the value will always be -1.

Demo

For the Leopard Speech-to-Text Web SDK, there is a Web demo project available on the Leopard Speech-to-Text GitHub repository.

Setup

Clone the Leopard Speech-to-Text repository from GitHub:

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

Usage

1. Install dependencies:

cd leopard/demo/react
npm install

2. Run the demo with the start script with a language code to start a local web server hosting the demo in the language of your choice (e.g. de -> German, es -> Spanish). To see a list of available languages, run start without a language code.

npm run start ${LANGUAGE}

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

Resources