logWMSE, an audio quality metric & loss function with support for digital silence target. Useful for training and evaluating audio source separation systems.

Variations of L1 SNR Loss function for training audio source separation machine learning models

Collection of community-made presets for PulseEffects tailored for TUXEDO laptops.

Unofficial PyTorch implementation of "Moises-Light: Resource-efficient Band-split U-Net For Music Source Separation"

The code for the MAPSS measures for source separation evaluation (ICLR, 2026)

Repository for subjective and objective evaluation of source separation algorithms

evalmedia is an open-source Python framework for evaluating the quality of AI-generated media (images, video, audio).

best stereo plugin for better discord

Create a Dolby.io Frontend Application To Display Waveform With Wavesurfer.js And Interactive Elements Using Konva In SolidJS

A small Windows utility to disable Bluetooth Hands-Free (HFP) mode. Prevents low-quality “headset” audio and keeps Bluetooth in stereo. Designed for games and apps that break sound quality.

Bulk upgrade your music library from lossy to lossless FLAC via Soulseek and SLSKD

MicLock fixes bad Mac Bluetooth headphone sound quality by preventing AirPods, Sony WH-1000XM and other headsets from becoming the default microphone.

Application to Measure the Quality of Audio Devices using GNU Radio

Next.js Starter Repo for an Educational Platform with Dolby.io

Scan audio RMS in seconds, offline. rms-scan runs FFmpeg under the hood to batch validate files, emit JSON reports, and fail fast when levels miss your threshold. Built for broadcast fixes, podcast QC, and pre-release gating.

Audio Analysis Tool - Real-time speech recognition, transcription and quality analysis with Vosk and SNR evaluation

iOS Audio Moderation Sample

Free audio quality analysis tool for the AI data industry. Quality scores, signal metrics, language detection, speaker diarization, and compliance checking.

Professional audio quality analyzer for Windows that scans and grades music libraries from S to F using FFmpeg-powered spectrum analysis. Detects upsampling artifacts, clipping, stereo phase issues, and evaluates loudness via EBU R128 with full batch processing and export support for CSV, JSON, and plain text reports.