P Le Cerf

A new variable frame analysis method for speech recognition

Variable frame rate (VFR) analysis is a technique used in speech processing and recognition for discarding frames that are too much alike. The article introduces a new method for VFR. Instead of calculating the distance between frames, the norm of the derivative parameters is used in deciding to retain or to discard a frame, informal inspection of speech spectrograms shows that this new method puts more emphasis on the transient regions of the speech signal. Experimental results with a hidden Markov model (HMM) based system show that the new method outperforms the classical method. >

Multilayer perceptrons as labelers for hidden Markov models

A novel combination of multilayer perceptrons (MLPs) and hidden Markov models (HMMs) is presented. Instead of using MLPs as probability generators for HMMs, the authors propose to use MLPs as labelers for discrete parameter HMMs. Compared with the probabilistic interpretation of MLPs, this gives them the advantage of flexibility in system design (e.g., the use of word models instead of phonetic models while using the same MLPs). Moreover, since they do not need to reach a global minimum, they can do with MLPs with fewer hidden nodes, which can be trained faster. In addition, they do not need to retrain the MLPs with segmentations generated by a Viterbi alignment. Compared with Euclidean labeling, their method has the advantages of needing fewer HMM parameters per state and obtaining a higher recognition accuracy. Several improvements of the baseline MLP labeling are investigated. When using one MLP, the best results are obtained when giving the labels a fuzzy interpretation. It is also possible to use parallel MLPs where each is based on a different parameter set (e.g., basic parameters, their time derivatives, and their second-order time derivatives). This strategy increases the recognition results considerably. A final improvement is the training of MLPs for subphoneme classification. >

Using parallel MLPs as labelers for multiple codebook HMMs

The authors investigate the use of multilayer perceptrons (MLPs) as labelers for a discrete parameter hidden Markov model (HMM) system. They introduce a number of strategies, of which the multi-MLP approach, which uses parallel MLPs for separate parameter sets, is the most promising. The performance of the new system is just as good as that of a classical discrete parameter HMM system (using multiple Euclidean vector quantization), but needs fewer HMM parameters (80 compared with 330 per state). Therefore, multi-MLP labeling is much more efficient than Euclidean labeling. >The authors investigate the use of multilayer perceptrons (MLPs) as labelers for a discrete parameter hidden Markov model (HMM) system. They introduce a number of strategies, of which the multi-MLP approach, which uses parallel MLPs for separate parameter sets, is the most promising. The performance of the new system is just as good as that of a classical discrete parameter HMM system (using multiple Euclidean vector quantization), but needs fewer HMM parameters (80 compared with 330 per state). Therefore, multi-MLP labeling is much more efficient than Euclidean labeling.<<ETX>>

Pseudo-segment based speech recognition using neural recurrent whole-word recognizers

Describes a recurrent neural network based, isolated word speech recognizer. The recognizer uses 2 MLPs. A first, static MLP is used for classification of frames in phonemes. Next, a time compression step is applied. The resulting pseudo-segments are then used as inputs for a second, dynamic MLP that integrates the information over time to decide the current word. The authors apply this approach on an isolated digit recognition task and compare the results with hybrid MLP/HMM approach using the same static MLP.<<ETX>>

Frame and frame dimension reduction techniques for automatic speech recognition

Concentrates on techniques for parameter and frame reduction. Standard parameter sets of contemporary speech recognition systems include some set of basic parameters (e.g. cepstra and energy), their time derivatives, and their second time derivatives. The large number of parameters (typically 25 to 45) induces the investigation of methods for reducing the amount of computation, without loss of recognition accuracy. The paper presents the variable frame rate analysis, a technique for leaving out frames that are too resemblant, and describes methods for decreasing the number of parameters in a frame. >Concentrates on techniques for parameter and frame reduction. Standard parameter sets of contemporary speech recognition systems include some set of basic parameters (e.g. cepstra and energy), their time derivatives, and their second time derivatives. The large number of parameters (typically 25 to 45) induces the investigation of methods for reducing the amount of computation, without loss of recognition accuracy. The paper presents the variable frame rate analysis, a technique for leaving out frames that are too resemblant, and describes methods for decreasing the number of parameters in a frame.<<ETX>>