Abstract-In this paper, we evaluate the vulnerability of speaker verification (SV) systems to synthetic speech. The SV systems are based on either the Gaussian mixture modeluniversal background model (GMM-UBM) or support vector machine (SVM) using GMM supervectors. We use a hidden Markov model (HMM)-based text-to-speech (TTS) synthesizer, which can synthesize speech for a target speaker using small amounts of training data through model adaptation of an average voice or background model. Although the SV systems have a very low equal error rate (EER), when tested with synthetic speech generated from speaker models derived from the Wall-Street Journal (WSJ) speech corpus, over 81% of the matched claims are accepted. This result suggests vulnerability in SV systems and thus a need to accurately detect synthetic speech. We propose a new feature based on relative phase shift (RPS), demonstrate reliable detection of synthetic speech, and show how this classifier can be used to improve security of SV systems.

In this paper, we present new results from our research into the vulnerability of a speaker verification (SV) system to synthetic speech. We use a HMM-based speech synthesizer, which creates synthetic speech for a targeted speaker through adaptation of a background model and both GMM-UBM and support vector machine (SVM) SV systems. Using 283 speakers from the Wall-Street Journal (WSJ) corpus, our SV systems have a 0.35% EER. When the systems are tested with synthetic speech generated from speaker models derived from the WSJ journal corpus, over 91% of the matched claims are accepted. We propose the use of relative phase shift (RPS) in order to detect synthetic speech and develop a GMM-based synthetic speech classifier (SSC). Using the SSC, we are able to correctly classify human speech in 95% of tests and synthetic speech in 88% of tests thus significantly reducing the vulnerability.

In the field of speaker verification (SV) it is nowadays feasible and relatively easy to create a synthetic voice to deceive a speech driven biometric access system. This paper presents a synthetic speech detector that can be connected at the front-end or at the back-end of a standard SV system, and that will protect it from spoofing attacks coming from stateof-the-art statistical Text to Speech (TTS) systems. The system described is a Gaussian Mixture Model (GMM) based binary classifier that uses natural and copy-synthesized signals obtained from the Wall Street Journal database to train the system models. Three different state-of-the-art vocoders are chosen and modeled using two sets of acoustic parameters: 1) relative phase shift and 2) canonical Mel Frequency Cepstral Coefficients (MFCC) parameters, as baseline. The vocoder dependency of the system and multivocoder modeling features are thoroughly studied. Additional phase-aware vocoders are also tested. Several experiments are carried out, showing that the phase-based parameters perform better and are able to cope with new unknown attacks. The final evaluations, testing synthetic TTS signals obtained from the Blizzard challenge, validate our proposal. IndexTerms-BIO-MODA-VOI, voice biometrics, anti-spoofing, phase information, synthetic speech detection.

No abstract

This paper describes a voice conversion system designed with the aim of improving the intelligibility and pleasantness of oesophageal voices. Two different systems have been built, one to transform the spectral magnitude and another one for the fundamental frequency, both based on DNNs. Ahocoder has been used to extract the spectral information (mel cepstral coefficients) and a specific pitch extractor has been developed to calculate the fundamental frequency of the oesophageal voices. The cepstral coefficients are converted by means of an LSTM network. The conversion of the intonation curve is implemented through two different LSTM networks, one dedicated to the voiced unvoiced detection and another one for the prediction of F0 from the converted cepstral coefficients. The experiments described here involve conversion from one oesophageal speaker to a specific healthy voice. The intelligibility of the signals has been measured with a Kaldi based ASR system. A preference test has been implemented to evaluate the subjective preference of the obtained converted voices comparing them with the original oesophageal voice. The results show that spectral conversion improves ASR while restoring the intonation is preferred by human listeners.