Neural Networks and Speech Processing

Morgan, D.P.; Scofield, Christopher L.

doi:10.1007/978-1-4615-3950-6_10

Cited by 18 publications

(20 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…A di!erent perspective [12] is to consider ANNs as optimal non-parametric estimators of Bayesian classi"ers. Finally, integration with the LM is sought whenever connectionist approaches to language modeling [85], in addition to acoustic modeling, are used: under these Table 4 Summary of main`other approachesa discussed in Section 3.5…”

Section: Some Future Issuesmentioning

confidence: 99%

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

2001

View full text Add to dashboard Cite

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.

show abstract

Section: Some Future Issuesmentioning

confidence: 99%

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

2001

View full text Add to dashboard Cite

show abstract

“…The signal modulation module perfectly matches our humidity sensor, flame sensor and temperature sensor, so it can solve commendably signal modulation problem of the whole system. [4,7] C. The problem of software reliability…”

Section: B Signal Modulationmentioning

confidence: 99%

“…High frequency filter is installed in the rectifying circuit to eliminate high frequency interference of the transformer and also protect the diode. 4 The modules, such as control computer, electronic control circuit, temperature control and test system, humidity control and test system, flame control system, are all supplied by single phase of AC. The adoption of this technique can further eliminate interference of the power supply.…”

Section: A Anti-interferencementioning

confidence: 99%

Design of the Intelligent Test Equipment of Airplane Fire Alarm System

Hengcheng

Zeng

Xiangqun

2010

2010 International Forum on Information Technology and Applications

View full text Add to dashboard Cite

The airplane fire alarm system is a piece of the most important equipment influencing the flight security. Due to the factors of design and working condition, the reliability of the system is poor, especially the high rate of false alarm. So the test equipment of airplane fire alarm system, an intelligent synthesis test system allowing multiple parameters and digital data acquisition, is developed. This system takes embedded industrial control computer as its core, and adopts unit wooden blocks type structure. The software, being programmed by C language, circulating in WINDOWS NT, is designed to manage the software and hardware resource of the entire system. This system has lots of advantages, such as high reliability and adaptability to the operating environment, fine commonality, easy to operate, and highly programmatic and automatic working process.

show abstract

“…This probability can then be used, after some modifications (Bourlard & Morgan, 1993), as local probabilities in HMMs. Morgan and Scofield (1991).…”

Section: Hybrid Hmm/ann Approachmentioning

confidence: 99%