TechPrufer
Sign inGet started

Blogs›Datadog›Client-Side Noise Suppression Library

Client-Side Noise Suppression Library

Client-Side Noise Suppression Library

1
posts
2025

Datadog developed and open-sourced dtln-rs, a real-time, client-side noise suppression library based on the Dual-Signal Transformation LSTM Network (DTLN). This library is designed to be embeddable in native clients and web applications, offering high performance without server dependencies. It leverages WebAssembly, Node.js native modules, and native Rust targets to integrate with WebRTC, enabling high-quality audio processing directly on the client device. The implementation focuses on optimizing deep learning models for efficient execution on standard hardware, addressing the limitations of existing off-the-shelf solutions and costly server-based approaches.

2025

How we built a real-time, client-side noise suppression library without server dependencies | Datadog

6/24/2025

This post details the creation and open-sourcing of dtln-rs, a Rust-based noise suppression library. It explains the motivation for building a client-side solution due to limitations of existing WebRTC noise reduction and server-dependent models. The post highlights the use of the DTLN algorithm, its implementation using TensorFlow Lite and Rust, and its packaging for WebAssembly, Node.js, and native clients. It emphasizes the performance optimizations achieved to enable real-time processing on client devices and the decision to open-source the library.