Developer Releases HERTZ: A 'Zero-Server' Browser Tool for Instant Vocal Polishing

Developer askpext has released HERTZ, a browser-based audio utility designed for "instant vocal polishing." The tool allows users to upload audio files, select specific regions, and apply a hardcoded signal chain to improve vocal clarity before exporting the result as an MP3.

To achieve this, HERTZ employs a "Zero-Server" architecture, ensuring that all digital signal processing occurs entirely on the client side. According to the project's GitHub documentation, the technical stack relies on the Web Audio API for real-time processing and Wavesurfer.js for waveform visualization and region selection. For the final export, the tool integrates FFmpeg.wasm to handle MP3 encoding directly within the browser. To enable the SharedArrayBuffer required by FFmpeg.wasm, the application necessitates specific security headers: Cross-Origin-Opener-Policy set to "same-origin" and Cross-Origin-Embedder-Policy set to "require-corp."

The Technical Signal Chain

Unlike full-featured Digital Audio Workstations (DAWs) that offer endless plugins, HERTZ implements a specific, streamlined vocal chain. The process begins with a High-Pass Filter featuring an 80Hz cutoff to remove low-end rumble. This is followed by a Dynamics Compressor with a threshold of -20dB and a ratio of 3.1:1 to stabilize volume levels. The chain concludes with Peak Normalization set to -3.0 dB to ensure the output is loud and consistent without clipping.

Shifting the Audio Workflow

Traditional audio editing typically requires heavy desktop software or cloud services that necessitate uploading sensitive files to remote servers. HERTZ bridges this gap by bringing a simplified "vocal polishing" workflow into the browser via WebAssembly (Wasm). This approach transforms the browser into a lightweight alternative for quick podcast or voice-over edits, removing the friction of software installation or server-side dependencies.

Implications for Browser-Based DSP

The project demonstrates the growing capability of modern browsers to handle complex digital signal processing (DSP) and file encoding tasks locally. By removing the need for server-side infrastructure, HERTZ enhances user privacy and eliminates the latency associated with uploading and downloading large audio files. It also significantly reduces operational costs for the developer.

Community Reception

Following its debut on Hacker News, the project drew attention for its simplicity and the nature of its development. One commenter described it as a "vibe-coded project," suggesting the developer likely used a Large Language Model (LLM) to implement the signal chain based on visual settings. While the commenter noted the trend of LLM-assisted coding, they praised the project for having source code that remains "fairly readable and small."