MUSAN

by JHU CLSP

109h corpus of music + speech + noise — augmentation backbone for ASR/SV.

TL;DR

109h corpus of music + speech + noise — augmentation backbone for ASR/SV.

Best for data augmentation (additive noise, music) for ASR + speaker verification training. Pricing: free.

Category
Open source
License
—
Stars
—
Last push
—
Pricing
free
Platforms
OpenSLR

What it is

MUSAN (OpenSLR-17) is the canonical augmentation corpus for speech systems — 109 hours of mixed speech, music, and noise, all CC BY 4.0.

Best for: Data augmentation (additive noise, music) for ASR + speaker verification training.
Watch out for: CC BY 4.0 · 60h speech + 42h music + 6h noise · curated to be permissively-licensed source material. Cite: Snyder et al., 2015.

Install / use

https://www.openslr.org/17/

Features

Speaker diarizationNo
Word-level timestampsNo
Streaming / real-timeNo
Languages supported0
HIPAA eligibleNo

MUSAN vs Whipscribe

FeatureMUSANWhipscribe
CategoryOpen sourceTranscription APIs
Pricingfreefree beta
Speaker diarizationNoYes
Word timestampsNoYes
StreamingNoNo
Languages099
PlatformsOpenSLRWeb, API, MCP

Alternatives to MUSAN

OpenAI Whisper
OpenAI

The reference open-source multilingual ASR model from OpenAI.

OSS · MIT ★ 98.1k
whisper.cpp
Georgi Gerganov

C/C++ port of Whisper — runs on anything, from a Raspberry Pi to Apple Silicon.

OSS · MIT ★ 48.8k
faster-whisper
SYSTRAN

4× faster than reference Whisper using CTranslate2 — production sweet spot.

OSS · MIT ★ 22.3k

Whipscribe is a managed faster-whisper + whisperX service. If you want transcripts without running infrastructure, paste a URL or drop a file in the form below — you'll have a transcript in seconds.