Noise-Robust Spoofed Speech Detection Using Discriminative Autoencoder
Open Access
- 29 June 2023
- journal article
- Published by Celal Bayar University Journal of Science in Celal Bayar Üniversitesi Fen Bilimleri Dergisi
- Vol. 19 (2) , 167-174
Abstract
Audio spoof detection gained attention of the researchers recently, as it is vital to detect spoofed speech for automatic speaker recognition systems. Publicly available datasets also accelerated the studies in this area. Many different features and classifiers have been proposed to overcome the spoofed speech detection problem, and some of them achieved considerably high performances. However, under additive noise, the spoof detection performance drops rapidly. On the other hand, number of studies about robust spoofed speech detection is very limited. The problem becomes more interesting as the conventional speech enhancement methods reportedly performed worse than no enhancement. In this work, i-vectors are used for spoof detection, and discriminative denoising autoencoder (DAE) network is used to obtain enhanced (clean) i-vectors from their noisy counterparts. Once the enhanced i-vectors are obtained, they can be treated as normal i-vectors and can be scored/classified without any modifications in the classifier part. Data from ASVspoof 2015 challenge is used with five different additive noise types, following a similar configuration of previous studies. The DAE is trained with a multicondition manner, using both clean and corrupted i-vectors. Three different noise types at various signal-to-noise ratios are used to create corrupted i-vectors, and two different noise types are used only in the test stage to simulate unknown noise conditions. Experimental results showed that the proposed DAE approach is more effective than the conventional speech enhancement methods.Keywords
Funding Information
- TÜBİTAK (121E057)
This publication has 30 references indexed in Scilit:
- Effectiveness of energy separation-based instantaneous frequency estimation for cochlear cepstral features for synthetic and voice-converted spoofed speech detectionComputer Speech & Language, 2022
- I-vector features and deep neural network modeling for language recognitionProcedia Computer Science, 2019
- Playback speech detection based on magnitude–phase spectrumElectronics Letters, 2018
- Fast i-vector denoising using MAP estimation and a noise distributions database for robust speaker recognitionComputer Speech & Language, 2017
- Constant Q cepstral coefficients: A spoofing countermeasure for automatic speaker verificationComputer Speech & Language, 2017
- Spoofing detection goes noisy: An analysis of synthetic speech detection in the presence of additive noiseSpeech Communication, 2016
- Additive noise compensation in the i-vector space for speaker recognition2015
- Unscented transform for ivector-based noisy speaker recognition2014
- Speaker adaptation of neural network acoustic models using i-vectors2013
- Assessment for automatic speech recognition: II. NOISEX-92: A database and an experiment to study the effect of additive noise on speech recognition systemsSpeech Communication, 1993