A DCT-based fast signal subspace technique for robust speech recognition

Automatic speech recognition (ASR) has gained wide popularity in last decade. Various devices like mobile phones, computers, vehicles, and audio/video players are now being equipped with ASR technology. The increasing use and dependence on ASR technology leads to research enhancements and opportunities in this domain. This chapter provides a detailed review of various advancements in ASR systems development. It highlights history of speech recognition followed by detailed insight into recent advancements and industry leaders providing latest solutions. ASR framework has been discussed in detail which includes feature extraction techniques, acoustic modeling techniques, and language modeling techniques. The chapter also lists various popular data sets available and discusses generation of new data sets. This work will be helpful for the researchers who are new to this field and are exploring development of new speech recognition techniques.

The authors propose here a new low‐frequency oscillatory (LFO) modes estimation scheme by applying Karhunen–Loeve transform (KLT) with total least square estimation of signal parameter employing rotational invariance technique (TLS‐ESPRIT). The proposed KLT provides the perfect decorrelation of the signal from the noise at low signal‐to‐noise ratio and for both white and coloured noise. KLT also maximally compress the energy of the estimated signal. KLT helps to estimate the signal subspace perfectly for any non‐stationary ambient power signal with the fastest computational time. First, the authors applied the KLT denoise to the filtered signal buried with highly correlated coloured Gaussian noise through the first eigenvectors of the correlation matrix and estimated the signal subspace. Finally, TLS‐ESPRIT is applied to estimate the modes of the power system. A comparative analysis of the KLT‐TLS‐ESPRIT with the modified TLS‐ESPRIT, multi‐channel wavelet transform, data driven‐stochastic subspace identification, TLS‐ESPRIT, Root‐MUSIC, and improved Prony method is presented. To explore the real‐time application of the KLT‐TLS‐ESPRIT, the authors considered Kundur's two‐area test system and 16‐machine 68‐bus system using 50,000 Monte‐Carlo simulations. The obtained results show that the KLT‐TLS‐ESPRIT accurately estimates the LFO with least mean and standard deviation as compared to the remaining six methods.

While a number of studies have investigated various speech enhancement and processing schemes for in-vehicle speech systems, little research has been performed using actual voice data collected in noisy car environments. We propose a new constrained switched adaptive beamforming algorithm (CSA-BF) for speech enhancement and recognition in real moving car environments. The proposed algorithm consists of a speech/noise constraint section, a speech adaptive beamformer, and a noise adaptive beamformer. We investigate CSA-BF performance with a comparison to classic delay-and-sum beamforming (DASB) in realistic car environments using a large quantity of data recorded in various car noise environments from across the United States. After analyzing the experimental results and considering the range of complex noise situations in the car environment using the CU-Move corpus, we formulate the CSA-BF algorithm. This method is shown to decrease WER (word error rate) for speech recognition by up to 31% and improve speech quality via the SEGSNR (segment signal-to-noise ratio) by up to 5.5 dB on the average, simultaneously.

This study is part of an ongoing collaborative effort between the medical and the signal processing communities to promote research on applying standard Automatic Speech Recognition (ASR) techniques for the automatic diagnosis of patients with severe obstructive sleep apnoea (OSA). Early detection of severe apnoea cases is important so that patients can receive early treatment. Effective ASR-based detection could dramatically cut medical testing time. Working with a carefully designed speech database of healthy and apnoea subjects, we describe an acoustic search for distinctive apnoea voice characteristics. We also study abnormal nasalization in OSA patients by modelling vowels in nasal and nonnasal phonetic contexts using Gaussian Mixture Model (GMM) pattern recognition on speech spectra. Finally, we present experimental findings regarding the discriminative power of GMMs applied to severe apnoea detection. We have achieved an 81% correct classification rate, which is very promising and underpins the interest in this line of inquiry.

Robust speech recognition in a vehicular environment has been an important research field that has attracted great research interest in recent years. Performance of general-purpose speech recognizers dramatically degrade because of various kinds of prevailing noise sources in cars. To deal with acoustic noises, we have proposed two noise reduction systems based on microphone array and post-filtering. In this chapter, we first describe these two noise reduction systems. Then, we present our investigation into the performance improvements of the automatic speech recognition (ASR) system for in-car applications when the two noise reduction systems are used as the front-end processors. We report the speech recognition experiments we have conducted using car noise recordings and the AURORA-2J speech database, as well as the recognition results we have obtained. Finally, based on these experimental results, we present some discussions on the proposed noise reduction systems.

Mobile technology is very fast growing and incredible, yet there are not much technology development and improvement for Deaf‐mute peoples. Existing mobile applications use sign language as the only option for communication with them. Before our article, no such application (app) that uses the disrupted speech of Deaf‐mutes for the purpose of social connectivity exists in the mobile market. The proposed application, named as vocalizer to mute (V2M), uses automatic speech recognition (ASR) methodology to recognize the speech of Deaf‐mute and convert it into a recognizable form of speech for a normal person. In this work mel frequency cepstral coefficients (MFCC) based features are extracted for each training and testing sample of Deaf‐mute speech. The hidden Markov model toolkit (HTK) is used for the process of speech recognition. The application is also integrated with a 3D avatar for providing visualization support. The avatar is responsible for performing the sign language on behalf of a person with no awareness of Deaf‐mute culture. The prototype application was piloted in social welfare institute for Deaf‐mute children. Participants were 15 children aged between 7 and 13 years. The experimental results show the accuracy of the proposed application as 97.9%. The quantitative and qualitative analysis of results also revealed that face‐to‐face socialization of Deaf‐mute is improved by the intervention of mobile technology. The participants also suggested that the proposed mobile application can act as a voice for them and they can socialize with friends and family by using this app.

When speech recognition is performed in a car environment, there are two important robustness issues that should be taken into account. The first robustness is related to the noisy acoustic condition, and it has been one of the most popular research topics of in-vehicle speech recognition. In contrast, the second robustness, which is related to unstable calibration of the audio input, has not attracted much attention. Consequently, the performance of speech recognition would degrade greatly in a real application if the input device such as a microphone array is badly calibrated. We propose robust feature combination in the MFCC domain using speech inputs from a linear microphone array. It realizes robust (from both the noise and the calibration viewpoints) and practical speech recognition applications in car environments. Even a simple MFCC averaging approach is effective, and a new algorithm, hypothesis-based feature combination (HBFC), improves the performance. We also extend cepstral variance normalization as variance re-scaling, which makes the feature combination approach more robust. The advantages of the proposed algorithms are confirmed by the experiments using the data recorded in a moving car.

Robust speech recognition in reverberant environments by using an optimal synthetic room impulse response model

While a number of studies have investigated various speech enhancement and processing schemes for in-vehicle speech systems, little research has been performed using actual voice data collected in noisy car environments. In this paper, we propose a new constrained switched adaptive beamforming algorithm (CSA-BF) for speech enhancement and recognition in real moving car environments. The proposed algorithm consists of a speech/noise constraint section, a speech adaptive beamformer, and a noise adaptive beamformer. We investigate CSA-BF performance with a comparison to classic delay-and-sum beamforming (DASB) in realistic car conditions using a corpus of data recorded in various car noise environments from across the U.S. After analyzing the experimental results and considering the range of complex noise situations in the car environment using the CU-Move corpus, we formulate the three specific processing stages of the CSA-BF algorithm. This method is evaluated and shown to simultaneously decrease word-error-rate (WER) for speech recognition by up to 31% and improve speech quality via the SEGSNR measure by up to +5.5 dB on the average.

This paper describes an attempt to extract peripheral features of a point c(t/sub i/,q/sub j/) on a time-quefrency (TQ) pattern by observing n/spl times/n neighborhoods of the point, and then to incorporate these peripheral features into the MFCC-based feature extractor of a speech recognition system as a replacement to dynamic features. In the design of the feature extractor, firstly, the orthogonal bases extracted directly from speech data by using the Karhunen-Loeve transform (KLT) of 7/spl times/3 blocks on a TQ pattern are adopted as the peripheral features, then, the upper two primal bases are selected and simplified in the form of /spl utri//sub t/-operator and /spl utri//sub q/-operator. The proposed feature-set of MFCC and peripheral features shows significant improvements in comparison with the standard feature-set of MFCC and dynamic features in experiments with an HMM-based automatic speech recognition (ASR) system. The reason for the increased performance is discussed in terms of minimal-pair tests.

Wiener filter is one of the most fundamental noise-reduction approaches among numerous techniques. The speech recognition in an in-vehicle environment needs a non-stationary noise cancellation to eliminate the background noise. However, few efforts have been reported to show the effectiveness of Wiener filter. Not much has been evaluated how the Wiener filter really works for reducing the non-stationary noise in real-time. In this paper, a real-time adaptive Wiener filter with two microphones is implemented to reduce noisy speech when noise signals and desired speech are incoming simultaneously. Furthermore, in order to build a real-time noise canceller, this paper also gives an analysis of different matrix sizes of the Wiener filter so as to enable the possibility of real-time implementation. The performance of the proposed design is measured by as much as 20dB noise reduction, and the proposed adaptive Wiener matrix update speed achieves a 28.6 ms/frame, with a matrix size of 200.

In this paper, we present a new technique for social mobile robot speech recognition based on scene-associated training models. The key contribution of the paper is a real-time framework that reduces the effect of room reverberation and ambient noise, a challenging problem in speech recognition. In classical approaches, anechoic sound is used to train the model, with the main focus on removing reverberation or noise from the sound. Our technique differs in that we train a number of speech recognizers directly from the reverberated sound, by associating each recognizer with a unique visual scene, to deal with the varying reverberation properties of different rooms. By extracting local features from a captured image and recognizing a scene, the robot can use the appropriate speech recognizer that is trained for the particular structural properties of that scene. We tested our method by using a baseline speech recognition model (HTK) across a variety of rooms and different levels of background noise. The results show that the association between a visual scene and a corresponding speech recognizer greatly improves the robot's speech recognition accuracy, together with increasing the computational speed of recognition, compared to competing techniques.

The histogram equalization approach is an efficient feature normalization technique for noise robust automatic speech recognition. However, it suffers from performance degradation when some fundamental conditions are not satisfied in the test environment. To remedy these limitations of the original histogram equalization methods, class-based histogram equalization approach has been proposed. Although this approach showed substantial performance improvement under noise environments, it still suffers from performance degradation due to the overfitting problem when test data are insufficient. To address this issue, the proposed histogram equalization technique employs the Bayesian estimation method in the test cumulative distribution function estimation. It was reported in a previous study conducted on the Aurora-4 task that the proposed approach provided substantial performance gains in speech recognition systems based on the acoustic modeling of the Gaussian mixture model-hidden Markov model. In this work, the proposed approach was examined in speech recognition systems with deep neural network-hidden Markov model (DNN-HMM), the current mainstream speech recognition approach where it also showed meaningful performance improvement over the conventional maximum likelihood estimation-based method. The fusion of the proposed features with the mel-frequency cepstral coefficients provided additional performance gains in DNN-HMM systems, which otherwise suffer from performance degradation in the clean test condition.

Automatic recognition of continuous speech for small vocabularies (e.g., telephone or credit card digit sequences) is possible with excellent accuracy, even in applications using telephone lines and serving a large population of users. However, even such simple recognition tasks suffer decreased performance in adverse conditions, e.g., significant background noise or fading on portable telephone channels. If we further impose significant limitations on the computing resources for the recognition task, then robust efficient speech recognition is still a significant challenge, even for simple vocabularies. Since many practical recognition tasks take place over telephone lines and in conditions that are less than optimal, speech recognition must be robust. The traditional hidden Markov model approach to speech recognition, using cepstral analysis, is computationally intensive and often does not work well under adverse acoustic conditions. We examine a simpler spectral analysis method, and suggest a segmental approach. High recognition accuracy can be maintained, despite a large decrease in both memory and computation, compared to traditional approaches.

In real-time environments an Automatic Speech Recognition (ASR) system in a car has to receive the driver’s voice only whilst suppressing background noise. Previous research by the authors presented a real-time Voice Activity Detector (VAD) which operates within a geometrical zone defined around the head of the desired speaker. The VAD only works well for a single sound source. During the period of desired speech the VAD does not work well if the unwanted speech is incoming at the same time as desired speech. As a solution to the above issue, this paper presents a real-time Wiener filter for noise reduction, improving the SNR by significant amounts in a car environment [1]. Experiments show that a simple template matching automatic speech recognition system obtained a 100% hit- rate [2].