Agent Voice Response - Resampler

A TypeScript npm package for audio resampling between different sample rates, specifically designed for AVR (Agent Voice Response) STS or TTS service integration.

Features

• Configurable Provider Sample Rate: Set the provider sample rate via constructor parameter
• Fixed Client Sample Rate: Client sample rate is fixed at 8000 Hz for compatibility (Asterisk AudioSocket Module)
• Bidirectional Resampling: Handle both upsampling (client → provider) and downsampling (provider → client)
• Buffered Processing: Accumulates audio chunks for optimal processing
• TypeScript Support: Full TypeScript support with type definitions

Installation

npm install avr-resampler

Usage

Basic Usage

import { AudioResampler } from 'avr-resampler';

const customResampler = new AudioResampler(48000); // 48 kHz provider rate
await customResampler.initialize();
const result = customResampler.downsample(audioBuffer);

Constructor Parameters

• providerSampleRate (number, optional): The sample rate of the audio provider in Hz. Defaults to 48000 Hz.

Methods

downsample(data: Buffer): Buffer

Processes incoming audio chunks from the provider and converts them to client format (8000 Hz).

• Input: Raw audio data from provider
• Output: Converted audio data if not enough data accumulated
• Usage: Call repeatedly with incoming audio chunks

upsample(data: Buffer): Buffer

Processes incoming audio chunks from the client (8000 Hz) and converts them to provider format.

• Input: Raw audio data from client
• Output: Converted audio data for provider
• Usage: Call with client audio chunks

destroy(): void

Cleans up resources and resets the resampler state.

initialize(): Promise<void>

Initializes the audio resamplers. Called automatically when needed.

Example

import { AudioResampler } from 'avr-resampler';

// Create resampler for 48 kHz provider
const resampler = new AudioResampler(48000);
await resampler.initialize();

// Process provider audio → client audio
const providerAudio = Buffer.from([/* your audio data */]);
const clientAudio = resampler.downsample(providerAudio);

if (clientAudio) {
  // Send to client
  sendToClient(clientAudio);
}

// Process client audio → provider audio
const clientInput = Buffer.from([/* client audio data */]);
const providerInput = await resampler.upsample(clientInput);

// Send to provider
sendToProvider(providerInput);

// Clean up when done
resampler.destroy();

Dependencies

• @alexanderolsen/libsamplerate-js: Audio resampling library

Support & Community

• GitHub: https://github.com/agentvoiceresponse - Report issues, contribute code.
• Discord: https://discord.gg/DFTU69Hg74 - Join the community discussion.
• Docker Hub: https://hub.docker.com/u/agentvoiceresponse - Find Docker images.
• NPM: https://www.npmjs.com/~agentvoiceresponse - Browse our packages.
• Wiki: https://wiki.agentvoiceresponse.com/en/home - Project documentation and guides.

Support AVR

AVR is free and open-source. If you find it valuable, consider supporting its development:

License

MIT License - see the LICENSE file for details.