2
$\begingroup$

I'm doing a student research project connected with VAD. And I want to find speech corpora with added noise. I found many references to the NIST Speaker Recognition Corpora. But unfortunately, all links on main site are already dead.

Maybe someone has this Corpora? Or someone can recommend a good Corpora with noise?

$\endgroup$
  • $\begingroup$ Welcome to DSP.SE! Most of those databases are paid for, you won't find them available for free. The licenses aren't too expensive, but they are non-zero cost. $\endgroup$ – Peter K. Apr 1 '16 at 15:11
  • $\begingroup$ You need corpus with speech or speech with noise. You could add noise on your own. $\endgroup$ – Gilles Apr 1 '16 at 17:38
1
$\begingroup$

Look into the MUSAN corpus from the Kaldi people (Dan Povey, et al):

MUSAN is a new corpus of music, speech, and noise. This dataset is suitable for training models for voice activity detection (VAD) and music/speech discrimination. Our corpus is released under a flexible Creative Commons license. The dataset consists of music from several genres, speech from twelve languages, and a wide assortment of technical and non-technical noises. We demonstrate use of this corpus for music/speech discrimination on Broadcast news and VAD for speaker identification.

$\endgroup$
0
$\begingroup$

Relevant paper is

"The QUT-NOISE-TIMIT corpus for the evaluation of voice activity detection algorithms"

http://eprints.qut.edu.au/38144/

Noise database is available here

https://www.qut.edu.au/research/research-projects/speech-audio-image-and-video-technology-saivt

For speech instead of TIMIT you can take any public database, for example, librispeech:

http://www.openslr.org/12/

$\endgroup$
0
$\begingroup$

Free speech dataset for VAD recorded in real environment with ground truth label:

refer: https://github.com/jtkim-kaist/VAD-Toolkit

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.