Julie Mauclair

Automatic Assessment of Speech Capability Loss in Disordered Speech

In this article, we report on the use of an automatic technique to assess pronunciation in the context of several types of speech disorders. Even if such tools already exist, they are more widely used in a different context, namely, Computer-Assisted Language Learning, in which the objective is to assess nonnative pronunciation by detecting learners’ mispronunciations at segmental and/or suprasegmental levels. In our work, we sought to determine if the Goodness of Pronunciation (GOP) algorithm, which aims to detect phone-level mispronunciations by means of automatic speech recognition, could also detect segmental deviances in disordered speech. Our main experiment is an analysis of speech from people with unilateral facial palsy. This pathology may impact the realization of certain phonemes such as bilabial plosives and sibilants. Speech read by 32 speakers at four different clinical severity grades was automatically aligned and GOP scores were computed for each phone realization. The highest scores, which indicate large dissimilarities with standard phone realizations, were obtained for the most severely impaired speakers. The corresponding speech subset was manually transcribed at phone level; 8.3% of the phones differed from standard pronunciations extracted from our lexicon. The GOP technique allowed the detection of 70.2% of mispronunciations with an equal rate of about 30% of false rejections and false acceptances. Finally, to broaden the scope of the study, we explored the correlation between GOP values and speech comprehensibility scores on a second corpus, composed of sentences recorded by six people with speech impairments due to cancer surgery or neurological disorders. Strong correlations were achieved between GOP scores and subjective comprehensibility scores (about 0.7 absolute). Results from both experiments tend to validate the use of GOP to measure speech capability loss, a dimension that could be used as a complement to physiological measures in pathologies causing speech disorders.

Audio indexing including frequency tracking of simultaneous multiple sources in speech and music

In this paper, we present a complete system for audio indexing. This system is based state-of-the-art methods of Speech-Music-Noise segmentation and Monophonic/Polyphonic estimation. After those methods we propose an original system of superposed sources detection. This approach is based on the analysis of the evolution of the predominant frequencies. In order to validate the whole system we used different corpora : Radio broadcasts, studio music and degraded field records. The first results are encouraging and show the potential of our approach which is generic and can be used on both music and speech contents.

Automatic identification of phonetic similarity based on underspecification

This paper presents a novel approach to the identification of phonetic similarity using properties observed during the speech recognition process. Experiments are presented whereby specific phones are removed during the training phase of a statistical speech recognition system so that the behaviour of the system can be analysed to see which alternative phone is selected. The domain of the analysis is restricted to specific contexts and the alternatively recognised (or substituted) phones are analysed with respect to a number of factors namely, the common phonetic properties, the phonetic neighbourhood and the frequency of occurrence with respect to a particular corpus. The results indicate that a measure of phonetic similarity based on alternatively recognised observed properties can be predicted based on a combination of these factors and as such can serve as an important additional source of information for the purposes of modelling pronunciation variation.

Pronunciation assessment of Japanese learners of French with GOP scores and phonetic information

In this paper, we report automatic pronunciation assessment experiments at phone-level on a read speech corpus in French, collected from 23 Japanese speakers learning French as a foreign language. We compare the standard approach based on Goodness Of Pronunciation (GOP) scores and phone-specific score thresholds to the use of logistic regressions (LR) models. French native speech corpus, in which artificial pronunciation errors were introduced, was used as training set. Two typical errors of Japanese speakers were considered: /o/ and /v/ of ten mispronounced as [l] and [b], respectively. The LR classifier achieved a 64.4% accuracy similar to the 63.8% accuracy of the baseline threshold method, when using GOP scores and the expected phone identity as input features only. A significant performance gain of 20.8% relative was obtained by adding phonetic and phonological features as input to the LR model, leading to a 77.1% accuracy. This LR model also outperformed another baseline approach based on linear discriminant models trained on raw f-BANK coefficient features.