audio-evaluation 开源项目推荐（16 个）

EvolvingLMMs-Lab /lmms-eval

适用于文本、图像、视频和音频任务的通用多模态评估工具包

wavlab-speech /versa

Versatile Evaluation of Speech and Audio

sp-nitech /listen-and-rate

A lightweight, self-hostable tool for conducting subjective listening tests in a browser.

OpenMOSS /MOSS-Audio-Tokenizer-Eval

An objective reconstruction evaluation toolkit for audio tokenizers, neural codecs, audio VAEs, and vocoders

ayusrjn /benten

Benten is an audio evaluation and analytics platform built specifically for Voice AI agents. It connects directly to voice AI platforms to discover your agents, process call audio, and measure conversation quality, response latency, and speech dynamics.

ruiling-smartbear /TIMMY

Transparent, reproducible, modular evaluation for AI-generated music

Tommieoxidative416 /video-evaluator

Extract grounded evidence from video files to enable automated review and visual understanding for coding agents.

mormegil6 /immersive-formats-evaluation

Supplementary materials (pipeline, data, R analysis) for a study of latency and anchor-choice confounds in objective binaural quality metrics (BAM-Q, BINAQUAL), evaluated across 11 immersive audio format variants and 3 professionally produced recordings.

AndersonMargalho /AndersonMargalho

Freelance Software Engineer | AI Automation | LLM Evaluation. Building open-source tools for AI testing.

riedemannai /Medical_ASR_Evaluator

A standalone tool for evaluating Automatic Speech Recognition (ASR) models, particularly optimized for medical/clinical speech recognition, using Word Error Rate (WER) metric

zyding123 /audio-analysis-app

macOS 音频批量分析工具：LUFS / LRA / 真峰值 / 削波检测 / 双口径频谱质心 / BPM + 频谱图表（PySide6 + ffmpeg + librosa）

pikaqiu2333 /DJBench

A reproducible benchmark for evaluating general-purpose agents on DJ mixing and real Mixxx software operation.

xiaotianming /music-eval

Local CPU workbench for evaluating and comparing AI-generated music with CLAP, SongEval and FAD

amelia-data /amelia-data

Professional portfolio for AI Evaluation, Audio Analysis, and UX Research. Specialized in Human-in-the-Loop (HITL) data integrity and design-focused research.

FajarIrnt /Profile-Portfolio

AI Data Annotation & Evaluation Portfolio featuring audio evaluation and image annotation projects.

nguyencuong102tg /vietnamese-ai-evaluation-sample

Sample Vietnamese AI evaluation, audio evaluation, and response ranking rubrics