Speaker recognition under mismatched conditions is a challenging task. Speech signal is nonlinear and nonstationary, and therefore, difficult to analyze under realistic conditions. Also, in real conditions, the nature of the noise present in speech data is not known a priori. In such cases, the performance of speaker identification or speaker verification degrades considerably under realistic conditions. Any SR system uses a voice activity detector as the front-end subsystem of the whole system. The performance of most VADs deteriorates at the (...) front end of the SR task or system under degraded conditions or in realistic conditions where noise plays a major role. Recently, speech data analysis and processing using Norden E. Huang’s empirical mode decomposition combined with Hilbert transform, commonly referred to as Hilbert–Huang transform, has become an emerging trend. EMD is an a posteriori, adaptive, data analysis tool used in time domain that is widely accepted by the research community. Recently, speech data analysis and speech data processing for speech recognition and SR tasks using EMD have been increasing. EMD-based VAD has become an important adaptive subsystem of the SR system that mostly mitigates the effect of mismatch between the training and the testing phase. Recently, we have developed a VAD algorithm using a zero-frequency filter-assisted peaking resonator and EMD. In this article, the efficacy of an EMD-based VAD algorithm is studied at the front end of a text-independent language-independent SI task for the speaker’s data collected in three languages at five different places, such as home, street, laboratory, college campus, and restaurant, under realistic conditions using EDIROL-R09 HR, a 24-bit wav/mp3 recorder. The performance of this proposed SI task is compared against the traditional energy-based VAD in terms of percentage identification rate. In both cases, widely accepted Mel frequency cepstral coefficients are computed by employing frame processing from the extracted voiced speech regions using the respective VAD techniques from the realistic speech utterances, and are used as a feature vector for speaker modeling using popular Gaussian mixture models. The experimental results showed that the proposed SI task with the VAD algorithm using ZFFPR and EMD at its front end performs better than the SI task with short-term energy-based VAD when used at its front end, and is somewhat encouraging. (shrink)

. Speaker recognition has been an active research area for many years. Methods to represent and quantify information embedded in speech signal are termed as features of the signal. The features are obtained, modeled and stored for further reference when the system is to be tested. Decision whether to accept or reject speakers are taken based on parameters of the data modeling techniques. Real world offers various degradations to the signal that hamper the signal quality. The degradations may be due (...) to ambient background noise, reverberation or multispeaker scenario. This paper presents a survey of various feature extraction, data modeling methods, metrics that are used to take the decisions and methods that can be used to preprocess the degraded data that have been used to perform the task of speaker recognition. (shrink)

This article presents an overview of the single-channel dereverberation methods suitable for distant speech recognition application. The dereverberation methods are mainly classified based on the domain of enhancement of speech signal captured by a distant microphone. Many single-channel speech enhancement methods focus on either denoising or dereverberating the distorted speech signal. There are very few methods that consider both noise and reverberation effects. Such methods are discussed under a multistage approach in this article. The article concludes with a hypothesis that (...) the methods that do not require an a priori reverberation impulse response is desirable in varying the environmental conditions for DSR applications such as intelligent home and office environments, humanoid robots, and automobiles rather than the methods that require an a priori reverberation impulse response. (shrink)

The performance of most of the state-of-the-art speaker recognition systems deteriorates under degraded conditions, owing to mismatch between the training and testing sessions. This study focuses on the front end of the speaker verification system to reduce the mismatch between training and testing. An adaptive voice activity detection algorithm using zero-frequency filter assisted peaking resonator was integrated into the front end of the SV system. The performance of this proposed SV system was studied under degraded conditions with 50 selected speakers (...) from the NIST 2003 database. The degraded condition was simulated by adding different types of noises to the original speech utterances. The different types of noises were chosen from the NOISEX-92 database to simulate degraded conditions at signal-to-noise ratio levels from 0 to 20 dB. In this study, widely used 39-dimension Mel frequency cepstral coefficient features were used, and Gaussian mixture model–universal background model was used for speaker modeling. The proposed system’s performance was studied against the energy-based VAD used as the front end of the SV system. The proposed SV system showed some encouraging results when EMD-based VAD was used at its front end. (shrink)

In this article, a new adaptive data-driven strategy for voice activity detection using empirical mode decomposition is proposed. Speech data are decomposed using an a posteriori, adaptive, data-driven EMD in the time domain to yield a set of physically meaningful intrinsic mode functions. Each IMF preserves the nonlinear and nonstationary property of the speech utterance. Among a set of IMFs, the IMF that contains source information dominantly called characteristic IMF can be identified and extracted by designing a zero-frequency filter-assisted peaking (...) resonator. The detected CIMF is used to compute energy using short-term processing. Choosing proper threshold, voiced regions in speech utterances are detected using frame energy. The proposed framework has been studied on both clean speech utterance and noisy speech utterance. The proposed method is used for voice activity detection in the presence of white noise and shows encouraging result in the presence of white noise up to 0 dB. (shrink)