Robust Feature Extraction Using Modulation Filtering of Autoregressive Models

Speaker recognition in noisy environments is challenging when there is a mis-match in the data used for enrollment and verification. In this paper, we propose a robust feature extraction scheme based on spectro-temporal modulation filtering using two-dimensional (2-D) autoregressive (AR) models. The first step is the AR modeling of the sub-band temporal envelopes by the application of the linear prediction on the sub-band discrete cosine transform (DCT) components. These sub-band envelopes are stacked together and used for a second AR modeling step. The spectral envelope across the sub-bands is approximated in this AR model and cepstral features are derived which are used for speaker recognition. The use of AR models emphasizes the focus on the high energy regions which are relatively well preserved in the presence of noise. The degree of modulation filtering is controlled using AR model order parameter. Experiments are performed using noisy versions of NIST 2010 speaker recognition evaluation (SRE) data with a stateof-art speaker recognition system. In these experiments, the proposed features provide significant improvements compared to baseline features (relative improvements of 20% in terms of equal error rate (EER) and 35 % in terms of miss rate at 10 % false alarm).

Speaker identification is a biometric technique of determining an unknown speaker's identity among a number of speakers using distinguish latent information of uttered speech. Crime investigation, security control, telephone banking and trading, and information reservation are some applications of this technique. Frequency Domain Linear Prediction (FDLP) is a time-frequency-based feature has been derived using 2-D autoregressive model. This feature was constructed from sub-bands short frame energies estimation. FDLP has been used in this study to propose a robust text-dependent speaker identification technique. The clean features were used to obtain speaker behavioural model. Support vector machine has been used to train the proposed method. This presented study was tested in both clean and noisy conditions to validate the method extensively. The proposed method got significant improved performance over all traditional methods performances in noisy conditions. The obtained performance was indicated; the proposed method was very robust to noises and showed consistent performance irrespective to noises.

Significance of analytic phase of speech signals in speaker verification

Local spectral variability features for speaker verification

Relevance factor of maximum a posteriori adaptation for GMM–NAP–SVM in speaker and language recognition

In this paper, we study a front-end frame selection approach for the interview channel speaker recognition system. This new approach keeps the high quality speech frames and removes noisy and irrelevant speech frames for speaker modeling. For robust voice activity detection (VAD) under the different types of microphones located in the interview room, we adopt the spectral subtraction algorithm for noise reduction. An energy based frame selection algorithm is first applied to indicate the speech activity at the frame level. To overcome the summed channel effects in the interview condition, a study is conducted to effectively extract the relevant speaker's speech frames based on VAD Tags and ASR transcript Tags provided by NIST. The eigenchannel based GMM-SVM speaker recognition system is used to evaluate the proposed method. The experiments are conducted on the NIST 2008 and NIST 2010 Speaker Recognition Evaluation interview-interview conditions. It demonstrates that the approach provides an efficient way to select high quality speech frames and the relevant speaker's voice in the interview environment for speaker recognition.

Speaker Recognition is the ability of the system to recognize the speaker from the set of speaker samples available in the system. It is of 2 types, one uses a keyword, called text-dependent systems, and another one can recognize the voice in any language/text, also called as text-independent speaker recognition. In this paper, a text-independent, language-independent speaker recognition system is implemented using dense & convolutional neural networks. Speaker recognition has found several applications in upcoming electronic products like personal/home assistants, telephone banking and biometric identification. In this paper, we explore a system that uses MFCC along with DNN and CNN as the model for building a speaker recognition system.

In this paper we propose softSAD: the direct integration of speech posteriors into a speaker recognition system as an alternative to using speech activity detection (SAD). Motivated by the need to use audio from short recordings more efficiently, softSAD removes the need to discard audio using speech/non-speech decisions based on a threshold as done with SAD. Instead, softSAD explicitly integrates into the Baum-Welch statistics a speech posterior for each frame. We compare softSAD and SAD in mismatched conditions by evaluating a system developed for the National Institute for Standards and Technology (NIST) 2012 speaker recognition evaluation (SRE) on the short test conditions of the channel-degraded Robust Automatic Transcription of Speech (RATS) speaker identification task (and vice versa). We demonstrate that softSAD provides benefit over SAD for short test audio in mismatched conditions.

Speech, languages, and background voices are overlapped in the multi-speaker environment. The speech separation process is necessary for downstream applications like language and speaker identification and ASR tasks. A combined speech separation and language recognition approach is proposed in this paper. First, the monolingual information is extracted from a multi-speaker multilingual speech scenario, then fed the separated speech to a language identification model. The speech separation model is trained in a single-channel overlapped speech data, and the language identification model is trained in Oriental language corpus. The combined effect of these two models is evaluated in the task of language identification. When the language recognition network is fed with the audio spectrograms separated by the source separation network, the Chinese and Korean language recognition results are significantly improved in terms of accuracy and recall rate than the mixed audio spectrograms.

The QUT-NOISE-SRE protocol is designed to mix the large QUT-NOISE database, consisting of over 10 hours of back- ground noise, collected across 10 unique locations covering 5 common noise scenarios, with commonly used speaker recognition datasets such as Switchboard, Mixer and the speaker recognition evaluation (SRE) datasets provided by NIST. By allowing common, clean, speech corpora to be mixed with a wide variety of noise conditions, environmental reverberant responses, and signal-to-noise ratios, this protocol provides a solid basis for the development, evaluation and benchmarking of robust speaker recognition algorithms, and is freely available to download alongside the QUT-NOISE database. In this work, we use the QUT-NOISE-SRE protocol to evaluate a state-of-the-art PLDA i-vector speaker recognition system, demonstrating the importance of designing voice-activity-detection front-ends specifically for speaker recognition, rather than aiming for perfect coherence with the true speech/non-speech boundaries.

Many powerful methods for speaker recognition have been introduced in recent years - high-level features, novel classifiers, and channel compensation methods. A common arena for evaluating these methods has been the NIST speaker recognition evaluation (SRE). In the NIST SRE from 2002-2005, a popular approach was to fuse multiple systems based upon cepstral features and different linguistic tiers of high-level features. With enough enrollment data, this approach produced dramatic error rate reductions and showed conceptually that better performance was attainable. A drawback in this approach is that many high-level systems were being run independently requiring significant computational complexity and resources. In 2006, MIT Lincoln Laboratory focused on a new system architecture which emphasized reduced complexity. This system was a carefully selected mixture of high-level techniques, new classifier methods, and novel channel compensation techniques. This new system has excellent accuracy and has substantially reduced complexity. The performance and computational aspects of the system are detailed on a NIST 2006 SRE task.

The recently introduced mean Hilbert envelope coefficients (MHEC) have been shown to be an effective alternative to MFCCs for robust speaker identification under noisy and reverberant conditions in relatively small tasks. In this study, we investigate the effectiveness of these acoustic features in the context of a state-of-the-art speaker recognition system. The i-vectors are used to represent the acoustic space of speakers, while modeling is performed via probabilistic linear discriminant analysis (PLDA). We report speaker verification performance on the NIST SRE-2010 extended telephone and microphone trials for both female and male genders. Experimental results confirm consistent superiority of MHECs to traditional MFCCs within i-vector speaker verification, particularly under microphone and telephone training-test mismatch conditions. In addition, fusion of subsystems trained with the individual front-ends proves that the two acoustic features (i.e., MHEC and MFCC) provide complimentary information for recognizing speakers.

In this paper, we compare several approaches for the extraction of modulation frequency features from speech signal using a phoneme recognition system. The general framework in these approaches is to decompose the speech signal into a set of sub-bands. Amplitude modulations (AM) in the sub-band signal are used to derive features for automatic speech recognition (ASR). Then, we propose a feature extraction technique which uses autoregressive models (AR) of sub-band Hilbert envelopes in relatively long segments of speech signal. AR models of Hilbert envelopes are derived using frequency domain linear prediction (FDLP). Features are formed by converting the FDLP envelopes into static and dynamic modulation frequency components. In the phoneme recognition experiments using the TIMIT database, the FDLP based modulation frequency features provide significant improvements compared to other techniques (average relative improvement of 7.5 % over the base-line features). Furthermore, a detailed analysis is performed to determine the relative contribution of various processing stages in the proposed technique.

Although the field of automatic speaker or speech recognition has been extensively studied over the past decades, the lack of robustness has remained a major challenge. The missing data technique (MDT) is a promising approach. However, its performance depends on the correlation across frequency bands. This paper presents a new reconstruction method for feature enhancement based on the trait. In this paper, the degree of concentration across frequency bands is measured with principal component analysis (PCA). Through theoretical analysis and experimental results, it is found that the correlation of the feature vector extracted from the sub-band (SB) is much stronger than the ones extracted from the full-band (FB). Thus, rather than dealing with the spectral features as a whole, this paper splits full-band into sub-bands and then individually reconstructs spectral features extracted from each SB based on MDT. At the end, those constructed features from all sub-bands will be recombined to yield the conventional mel-frequency cepstral coefficient (MFCC) for recognition experiments. The 2-sub-band reconstruction approach is evaluated in speaker recognition system. The results show that the proposed approach outperforms full-band reconstruction in terms of recognition performance in all noise conditions. Finally, we particularly discuss the optimal selection of frequency division ways for the recognition task. When FB is divided into much more sub-bands, some of the correlations across frequency channels are lost. Consequently, efficient division ways need to be investigated to perform further recognition performance.

Investigation of the effect of data duration and speaker gender on text-independent speaker recognition