Multiple cameras for audio-visual speech recognition in an automotive environment

Audio-visualspeechrecognition, or the combination of visual lip-reading with traditional acoustic speechrecognition, has been previously shown to provide a considerable improvement over acoustic-only approaches in noisy environments, such as that present in an automotive cabin. The research presented in this paper will extend upon the established audio-visualspeechrecognition literature to show that further improvements in speechrecognition accuracy can be obtained when multiple frontal or near-frontal views of a speaker's face are available. A series of visualspeechrecognition experiments using a four-stream visual synchronous hidden Markov model (SHMM) are conducted on the four-camera AVICAR automotiveaudio-visualspeech database. We study the relative contribution between the side and central orientated cameras in improving visualspeechrecognition accuracy. Finally combination of the four visual streams with a single audio stream in a five-stream SHMM demonstrates a relative improvement of over 56% in word recognition accuracy when compared to the acoustic-only approach in the noisiest conditions of the AVICAR database.

A Likelihood-Maximizing Framework for Enhanced In-Car Speech Recognition Based on Speech Dialog System Interaction

Speech recognition in car environments has been identified as a valuable means for reducing driver distraction when operating noncritical in-car systems. Under such conditions, however, speech recognition accuracy degrades significantly, and techniques such as speech enhancement are required to improve these accuracies. Likelihood-maximizing (LIMA) frameworks optimize speech enhancement algorithms based on recognized state sequences rather than traditional signal-level criteria such as maximizing signal-to-noise ratio. LIMA frameworks typically require calibration utterances to generate optimized enhancement parameters that are used for all subsequent utterances. Under such a scheme, suboptimal recognition performance occurs in noise conditions that are significantly different from that present during the calibration session – a serious problem in rapidly changing noise environments out on the open road. In this chapter, we propose a dialog-based design that allows regular optimization iterations in order to track the ever-changing noise conditions. Experiments using Mel-filterbank noise subtraction (MFNS) are performed to determine the optimization requirements for vehicular environments and show that minimal optimization is required to improve speech recognition, avoid over-optimization, and ultimately assist with semireal-time operation. It is also shown that the proposed design is able to provide improved recognition performance over frameworks incorporating a calibration session only.

Channel selection in the short-time modulation domain for distant speech recognition

Automatic speech recognition from multiple distant micro- phones poses significant challenges because of noise and reverberations. The quality of speech acquisition may vary between microphones because of movements of speakers and channel distortions. This paper proposes a channel selection approach for selecting reliable channels based on selection criterion operating in the short-term modulation spectrum domain. The proposed approach quantifies the relative strength of speech from each microphone and speech obtained from beamforming modulations. The new technique is compared experimentally in the real reverb conditions in terms of perceptual evaluation of speech quality (PESQ) measures and word error rate (WER). Overall improvement in recognition rate is observed using delay-sum and superdirective beamformers compared to the case when the channel is selected randomly using circular microphone arrays.

Multi Pattern Dynamic Time Warping For Automatic Speech Recognition

We are addressing the problem of jointly using multiple noisy speech patterns for automatic speech recognition (ASR), given that they come from the same class. If the user utters a word K times, the ASR system should try to use the information content in all the K patterns of the word simultaneously and improve its speech recognition accuracy compared to that of the single pattern based speech recognition. T address this problem, recently we proposed a Multi Pattern Dynamic Time Warping (MPDTW) algorithm to align the K patterns by finding the least distortion path between them. A Constrained Multi Pattern Viterbi algorithm was used on this aligned path for isolated word recognition (IWR). In this paper, we explore the possibility of using only the MPDTW algorithm for IWR. We also study the properties of the MPDTW algorithm. We show that using only 2 noisy test patterns (10 percent burst noise at -5 dB SNR) reduces the noisy speech recognition error rate by 37.66 percent when compared to the single pattern recognition using the Dynamic Time Warping algorithm.

Joint evaluation of multiple speech patterns for speech recognition and training

We are addressing the novel problem of jointly evaluating multiple speech patterns for automatic speech recognition and training. We propose solutions based on both the non-parametric dynamic time warping (DTW) algorithm, and the parametric hidden Markov model (HMM). We show that a hybrid approach is quite effective for the application of noisy speech recognition. We extend the concept to HMM training wherein some patterns may be noisy or distorted. Utilizing the concept of ``virtual pattern'' developed for joint evaluation, we propose selective iterative training of HMMs. Evaluating these algorithms for burst/transient noisy speech and isolated word recognition, significant improvement in recognition accuracy is obtained using the new algorithms over those which do not utilize the joint evaluation strategy.

Joint decoding of multiple speech patterns for robust speech recognition

We are addressing a new problem of improving automatic speech recognition performance, given multiple utterances of patterns from the same class. We have formulated the problem of jointly decoding K multiple patterns given a single Hidden Markov Model. It is shown that such a solution is possible by aligning the K patterns using the proposed Multi Pattern Dynamic Time Warping algorithm followed by the Constrained Multi Pattern Viterbi Algorithm The new formulation is tested in the context of speaker independent isolated word recognition for both clean and noisy patterns. When 10 percent of speech is affected by a burst noise at -5 dB Signal to Noise Ratio (local), it is shown that joint decoding using only two noisy patterns reduces the noisy speech recognition error rate to about 51 percent, when compared to the single pattern decoding using the Viterbi Algorithm. In contrast a simple maximization of individual pattern likelihoods, provides only about 7 percent reduction in error rate.

Comparison of AM-FM Based Features For Robust Speech Recognition

Effective feature extraction for robust speech recognition is a widely addressed topic and currently there is much effort to invoke non-stationary signal models instead of quasi-stationary signal models leading to standard features such as LPC or MFCC. Joint amplitude modulation and frequency modulation (AM-FM) is a classical non-parametric approach to non-stationary signal modeling and recently new feature sets for automatic speech recognition (ASR) have been derived based on a multi-band AM-FM representation of the signal. We consider several of these representations and compare their performances for robust speech recognition in noise, using the AURORA-2 database. We show that FEPSTRUM representation proposed is more effective than others. We also propose an improvement to FEPSTRUM based on the Teager energy operator (TEO) and show that it can selectively outperform even FEPSTRUM

Fast Likelihood Computation in Speech Recognition using Matrices

Acoustic modeling using mixtures of multivariate Gaussians is the prevalent approach for many speech processing problems. Computing likelihoods against a large set of Gaussians is required as a part of many speech processing systems and it is the computationally dominant phase for Large Vocabulary Continuous Speech Recognition (LVCSR) systems. We express the likelihood computation as a multiplication of matrices representing augmented feature vectors and Gaussian parameters. The computational gain of this approach over traditional methods is by exploiting the structure of these matrices and efficient implementation of their multiplication. In particular, we explore direct low-rank approximation of the Gaussian parameter matrix and indirect derivation of low-rank factors of the Gaussian parameter matrix by optimum approximation of the likelihood matrix. We show that both the methods lead to similar speedups but the latter leads to far lesser impact on the recognition accuracy. Experiments on 1,138 work vocabulary RM1 task and 6,224 word vocabulary TIMIT task using Sphinx 3.7 system show that, for a typical case the matrix multiplication based approach leads to overall speedup of 46 % on RM1 task and 115 % for TIMIT task. Our low-rank approximation methods provide a way for trading off recognition accuracy for a further increase in computational performance extending overall speedups up to 61 % for RM1 and 119 % for TIMIT for an increase of word error rate (WER) from 3.2 to 3.5 % for RM1 and for no increase in WER for TIMIT. We also express pairwise Euclidean distance computation phase in Dynamic Time Warping (DTW) in terms of matrix multiplication leading to saving of approximately of computational operations. In our experiments using efficient implementation of matrix multiplication, this leads to a speedup of 5.6 in computing the pairwise Euclidean distances and overall speedup up to 3.25 for DTW.

Gammatone Wavelet Cepstral Coefficients for Robust Speech Recognition

We develop noise robust features using Gammatone wavelets derived from the popular Gammatone functions. These wavelets incorporate the characteristics of human peripheral auditory systems, in particular the spatially-varying frequency response of the basilar membrane. We refer to the new features as Gammatone Wavelet Cepstral Coefficients (GWCC). The procedure involved in extracting GWCC from a speech signal is similar to that of the conventional Mel-Frequency Cepstral Coefficients (MFCC) technique, with the difference being in the type of filterbank used. We replace the conventional mel filterbank in MFCC with a Gammatone wavelet filterbank, which we construct using Gammatone wavelets. We also explore the effect of Gammatone filterbank based features (Gammatone Cepstral Coefficients (GCC)) for robust speech recognition. On AURORA 2 database, a comparison of GWCCs and GCCs with MFCCs shows that Gammatone based features yield a better recognition performance at low SNRs.

Exploiting Chinese character models to improve speech recognition performance

The Chinese language is based on characters which are syllabic in nature. Since languages have syllabotactic rules which govern the construction of syllables and their allowed sequences, Chinese character sequence models can be used as a first level approximation of allowed syllable sequences. N-gram character sequence models were trained on 4.3 billion characters. Characters are used as a first level recognition unit with multiple pronunciations per character. For comparison the CU-HTK Mandarin word based system was used to recognize words which were then converted to character sequences. The character only system error rates for one best recognition were slightly worse than word based character recognition. However combining the two systems using log-linear combination gives better results than either system separately. An equally weighted combination gave consistent CER gains of 0.1-0.2% absolute over the word based standard system. Copyright © 2009 ISCA.