We formulate a novel approach to estimating the parameters of continuous density HMMs for speaker-independent (SI) continuous speech recognition. It is motivated by the fact that variability in SI acoustic models is attributed to both phonetic variation and variation among the speakers of the training population, that is independent of the information content of the speech signal. These two variation sources are decoupled and the proposed method jointly annihilates the inter-speaker variation and estimates the HMM parameters of the SI acoustic models. We compare the proposed training algorithm to the common SI training paradigm within the context of supervised adaptation. We show that the proposed acoustic models are more efficiently adapted to the test speakers, thus achieving significant overall word error rate reductions of 19% and 25% for 20K and 5K vocabulary tasks respectively.

/pdf/a-compact-model-for-speaker-adaptive-training-1iz8lh30ku.pdf

A compact model for speaker-adaptive training

https://hal.archives-ouvertes.fr/hal-00499180/document

Automatic speech recognition and speech variability: A review

In an effort to reduce the degradation in speech recognition performance caused by variation in vocal tract shape among speakers, a frequency warping approach to speaker normalization is investigated. A set of low complexity, maximum likelihood based frequency warping procedures have been applied to speaker normalization for a telephone based connected digit recognition task. This paper presents an efficient means for estimating a linear frequency warping factor and a simple mechanism for implementing frequency warping by modifying the filterbank in mel-frequency cepstrum feature analysis. An experimental study comparing these techniques to other well-known techniques for reducing variability is described. The results have shown that frequency warping is consistently able to reduce word error rate by 20% even for very short utterances.

/pdf/a-frequency-warping-approach-to-speaker-normalization-56s3hvnfo5.pdf

A frequency warping approach to speaker normalization

To advance research, it is important to identify promising future research directions, especially those that have not been adequately pursued or funded in the past. The working group producing this article was charged to elicit from the human language technology (HLT) community a set of well-considered directions or rich areas for future research that could lead to major paradigm shifts in the field of automatic speech recognition (ASR) and understanding. ASR has been an area of great interest and activity to the signal processing and HLT communities over the past several decades. As a first step, this group reviewed major developments in the field and the circumstances that led to their success and then focused on areas it deemed especially fertile for future research. Part 1 of this article will focus on historically significant developments in the ASR area, including several major research efforts that were guided by different funding agencies, and suggest general areas in which to focus research.

/pdf/developments-and-directions-in-speech-recognition-and-1g7hryudwz.pdf

Developments and directions in speech recognition and understanding, Part 1 [DSP Education]

Current automatic speech recognition systems make use of a single source of information about their input, viz a preprocessed form of the acoustic speech signal, which encodes the time-frequency distribution of signal energy The goal of this thesis is to investigate the benefits of integrating articulatory information into state-of-the art speech recognizers, either as a genuine alternative to standard acoustic representations, or as an additional source of information Articulatory information is represented in terms of abstract articulatory classes or "features", which are extracted from the speech signal by means of statistical classifiers A higher-level classifier then combines the scores for these features and maps them to standard subword unit probabilities
The main motivation for this approach is to improve the robustness of speech recognition systems in adverse acoustic environments, such as background noise Typically, recognition systems show a sharp decline of performance under these conditions We argue and demonstrate empirically that the articulatory feature approach can lead to greater robustness by enhancing the accuracy of the bottom-up acoustic modeling component in a speech recognition system
The second focus point of this thesis is to provide detailed analyses of the different types of information provided by the acoustic and the articulatory representations, respectively, and to develop strategies to optimally combine them To this effect we investigate combination methods at the levels of feature extraction, subword unit probability estimation, and word recognition
The feasibility of this approach is demonstrated with respect to two different speech recognition tasks The first of these is an American English corpus of telephone-bandwidth speech; the recognition domain is continuous numbers The second is a German database of studio-quality speech consisting of spontaneous dialogues In both cases recognition performance will be tested not only under clean acoustic conditions but also under deteriorated conditions

/pdf/robust-speech-recognition-using-articulatory-information-32o2izd983.pdf

Robust speech recognition using articulatory information

Differences in vocal tract size among individual speakers contribute to the variability of speech waveforms. The first-order effect of a difference in vocal tract length is a scaling of the frequency axis; a female speaker, for example, exhibits formants roughly 20% higher than the formants of from a male speaker, with the differences most severe in open vocal tract configurations. We describe a parametric method of normalisation which counteracts the effect of varied vocal tract length. The method is shown to be effective across a wide range of recognition systems and paradigms, but is particularly helpful in the case of a small amount of training data.

E. Eide

Papers

A parametric approach to vocal tract length normalization