Gwen Van Nuffelen

Automated intelligibility assessment of pathological speech using phonological features

It is commonly acknowledged that word or phoneme intelligibility is an important criterion in the assessment of the communication efficiency of a pathological speaker. People have therefore put a lot of effort in the design of perceptual intelligibility rating tests. These tests usually have the drawback that they employ unnatural speech material (e.g., nonsense words) and that they cannot fully exclude errors due to listener bias. Therefore, there is a growing interest in the application of objective automatic speech recognition technology to automate the intelligibility assessment. Current research is headed towards the design of automated methods which can be shown to produce ratings that correspond well with those emerging from a well-designed and well-performed perceptual test. In this paper, a novel methodology that is built on previous work (Middag et al., 2008) is presented. It utilizes phonological features, automatic speech alignment based on acoustic models that were trained on normal speech, context-dependent speaker feature extraction, and intelligibility prediction based on a small model that can be trained on pathological speech samples. The experimental evaluation of the new system reveals that the root mean squared error of the discrepancies between perceived and computed intelligibilities can be as low as 8 on a scale of 0 to 100.

Speech technology-based assessment of phoneme intelligibility in dysarthria

BACKGROUNDnCurrently, clinicians mainly rely on perceptual judgements to assess intelligibility of dysarthric speech. Although often highly reliable, this procedure is subjective with a lot of intrinsic variables. Therefore, certain benefits can be expected from a speech technology-based intelligibility assessment. Previous attempts to develop an automated intelligibility assessment mainly relied on automatic speech recognition (ASR) systems that were trained to recognize the speech of persons without known impairments. In this paper automatic speech alignment (ASA) systems are used instead. In addition, previous attempts only made use of phonemic features (PMF). However, since articulation is an important contributing factor to intelligibility of dysarthric speech and since phonological features (PLF) are shared by multiple phonemes, phonological features may be more appropriate to characterize and identify dysarthric phonemes.nnnAIMSnTo investigate the reliability of objective phoneme intelligibility scores obtained by three types of intelligibility models: models using only phonemic features (yielded by an automated speech aligner) (PMF models), models using only phonological features (PLF models), and models using a combination of phonemic and phonological features (PMF + PLF models).nnnMETHODS & PROCEDURESnCorrelations were calculated between the objective phoneme intelligibility scores of 60 dysarthric speakers and the corresponding perceptual phoneme intelligibility scores obtained by a standardized perceptual phoneme intelligibility assessment.nnnOUTCOMES & RESULTSnThe correlations between the objective and perceptual intelligibility scores range from 0.793 for the PMF models, over 0.828 for PLF models to 0.943 for PMF + PLF models. The features selected to obtain such high correlations can be divided into six main subgroups: (1) vowel-related phonemic and phonological features, (2) lateral-related features, (3) silence-related features, (4) fricative-related features, (5) velar-related features and (6) plosive-related features.nnnCONCLUSIONS & IMPLICATIONSnThe phoneme intelligibility scores of dysarthric speakers obtained by the three investigated intelligibility model types are reliable. The highest correlation between the perceptual and objective intelligibility scores was found for models combining phonemic and phonological features. The intelligibility scoring system is now ready to be implemented in a clinical tool.

Medialization thyroplasty: vocal outcome of silicone and titanium implant

Unilateral vocal fold paralysis can result in incomplete glottic closure, compromising the vocal efficiency and increasing the risk of aspiration. The glottic gap can be reduced by medialization thyroplasty. Both silicone and titanium implants have been described to be efficient in the improvement of the vocal outcome. The aim of this study was to assess the vocal outcome after medialization thyroplasty using silicone and titanium implants and to compare the results of the two implanted materials using prospective sequential cohort study. 24 patients underwent medialization thyroplasty, performed under local anaesthesia with intravenous sedation: 10 patients were included in the silicone cohort and 14 in the titanium cohort. All patients underwent vocal assessment preoperatively and postoperatively, using the Voice Handicap Index, the GRBASI scale, maximum phonation time and the Dysphonia Severity Index. Postoperative analysis of the entire population showed statistically significant improvement for the Voice Handicap Index, maximum phonation time, Dysphonia Severity Index and the parameters G, B and A of the GRBASI scale (83% follow up). Subgroup analysis showed a statistically significant greater improvement of Voice Handicap Index of the titanium cohort compared with the silicone cohort. Improvement of maximum phonation time, Dysphonia Severity Index and GRBASI scale of the titanium cohort was greater than improvement of the silicone cohort, but this difference was not statistically significant. Both silicone and titanium medialization implants showed improvement of vocal outcome, the results of the titanium implant being superior.

Study protocol for a randomized controlled trial: tongue strengthening exercises in head and neck cancer patients, does exercise load matter?

BackgroundReduced tongue strength is an important factor contributing to early and late dysphagia in head and neck cancer patients previously treated with chemoradiotherapy. The evidence is growing that tongue strengthening exercises can improve tongue strength and swallowing function in both healthy and dysphagic subjects. However, little is known about the impact of specific features of an exercise protocol for tongue strength on the actual outcome (strength or swallowing function). Previous research originating in the fields of sports medicine and physical rehabilitation shows that the degree of exercise load is an influential factor for increasing muscle strength in the limb skeletal muscles. Since the tongue is considered a muscular hydrostat, it remains to be proven whether the same concepts will apply.Methods/DesignThis ongoing randomized controlled trial in chemoradiotherapy-treated patients with head and neck cancer investigates the effect of three tongue strengthening exercise protocols, with different degrees of exercise load, on tongue strength and swallowing. At enrollment, 51 patients whose dysphagia is primarily related to reduced tongue strength are randomly assigned to a training schedule of 60, 80, or 100 % of their maximal tongue strength. Patients are treated three times a week for 8 weeks, executing 120 repetitions of the assigned exercise once per training day. Exercise load is progressively adjusted every 2 weeks. Patients are evaluated before, during and after treatment by means of tongue strength measurements, fiber-optic endoscopic evaluation of swallowing and quality-of-life questionnaires.DiscussionThis randomized controlled trial is the first to systematically investigate the effect of different exercise loads in tongue strengthening exercise protocols. The results will allow the development of more efficacious protocols.Trial registrationCurrent Controlled Trials ISRCTN14447678.