A survey of hybrid ANN/HMM models for automatic speech recognition

Abstract

In spite of the advances accomplished throughout the last decades, automatic speech recognition (ASR) is still a challenging and di$cult task. In particular, recognition systems based on hidden Markov models (HMMs) are e!ective under many circumstances, but do su!er from some major limitations that limit applicability of ASR technology in real-world environments. Attempts were made to overcome these limitations with the adoption of arti"cial neural networks (ANN) as an alternative paradigm for ASR, but ANN were unsuccessful in dealing with long time-sequences of speech signals. Between the end of the 1980s and the beginning of the 1990s, some researchers began exploring a new research area, by combining HMMs and ANNs within a single, hybrid architecture. The goal in hybrid systems for ASR is to take advantage from the properties of both HMMs and ANNs, improving #exibility and recognition performance. A variety of di!erent architectures and novel training algorithms have been proposed in literature. This paper reviews a number of signi"cant hybrid models for ASR, putting together approaches and techniques from a highly specialistic and non-homogeneous literature. E!orts concentrate on describing and referencing architectures and algorithms, their advantages and limitations, as well as on categorizing them into broad classes. Early attempts to emulate HMMs by ANNs are "rst described. Then we focus on ANNs to estimate posterior probabilities of the states of an HMM and on`globala optimization, where a single, overall training criterion is de"ned over the HMM and the ANNs. Connectionist vector quantization for discrete HMMs, and other more recent approaches are also reviewed. It is pointed out that, in addition to their theoretical interest, hybrid systems have been allowing for tangible improvements in recognition performance over the standard HMMs in di$cult and signi"cant benchmark tasks.

Robust ASR using Support Vector Machines

“…In later paragraphs, we briefly describe some of the most relevant ones. A complete survey about this subject can be found in [20].…”

Metaheuristic design of feedforward neural networks: A review of two decades of research

“…Therefore, FNN architecture simplification or model's complexity reduction is a challenging task. Such problem can be addressed through the integration of FNN with statistical methods like the one usually done with hidden Markov model [348]. Therefore, such kind of modification to network architecture and specialized node design may lead to different paradigms of FNN that may solve various real-world complex problems.…”

Sistema audiovisual para reconocimiento de comandos

“…El problema de reconocimiento automático del habla (ASR-"Audio Speech Recognition") mediante el análisis de señales de audio ha sido ampliamente abordado en la literatura [15,19]. Reconocimiento del habla basado en fonemas La aproximación más natural en sistemas ASR es usar fonemas como unidades básicas que conforman el habla, ya que de esta forma los humanos reconocemos la información contenida en la señal de voz, concatenando sonidos que forman las palabras; siendo capaces además de ignorar aquellos sonidos que no conllevan a una respuesta lógica y separar palabras que unimos cuando hablamos de forma continua.…”

“…El problema de reconocimiento automático del habla en señales de audio se ha tratado regularmente a través del modelado de las señales, utilizando técnicas como Redes Neuronales [14] o Modelos Ocultos de Markov [19], las cuales reportan buenos resultados en la literatura. Sin embargo, cuando las condiciones acústicas son adversas, su desempeño se ve afectado.…”