No abstract

No abstract

Background: The articulatory accuracy of patients with dysarthria is one of the most affected speech dimensions with a high impact on speech intelligibility. Behavioural treatments of articulation can either involve direct or indirect approaches. The latter have been thoroughly investigated and are generally appreciated for their almost immediate effects on articulation and intelligibility. The number of studies on (short-term) direct articulation therapy is limited. Aims: To investigate the effects of short-term, boost articulation therapy (BArT) on speech intelligibility in patients with chronic or progressive dysarthria and the effect of severity of dysarthria on the outcome. Methods & Procedures: The study consists of a two-group pre-/post-test design to assess speech intelligibility at phoneme and sentence level and during spontaneous speech, automatic speech and reading a phonetically balanced text. A total of 17 subjects with mild to severe dysarthria participated in the study and were randomly assigned to either a patient-tailored, intensive articulatory drill programme or an intensive minimal pair training. Both training programmes were based on the principles of motor learning. Each training programme consisted of five sessions of 45 min completed within one week. Outcomes & Results: Following treatment, a statistically significant increase of mean group intelligibility was shown at phoneme and sentence level, and in automatic sequences. This was supported by an acoustic analysis that revealed a reduction in formant centralization ratio. Within specific groups of severity, large and moderate positive effect sizes with Cohen's d were demonstrated. Conclusions & Implications: BArT successfully improves speech intelligibility in patients with chronic or progressive dysarthria at different levels of the impairment.

No abstract

In this paper a critical review of some reported algorithms is made, regarding the objective measurement of features in pathologic speech. The discussion is grouped in the following topics: prosody, pith related measurements, hoarseness and vowel/consonant production. The prosody topic includes measurements of timing, stress, tempo and intonation. Pitch related measurements involve the algorithms based on pitch period, hoarseness groups measurements of noise and turbulence related to the pharyngeal functioning, while vowel/consonant production includes formant’s transitions, voice onset time and nasality perturbations. Every algorithm was analyzed taking into account the following aspects: Domain of analysis, sampling frequency, sample of speech or voice used, amount of patients and kind of disease, correlation with subjective evaluations. Limitations of analyzed algorithms are shown and the desired characteristics for each topic’s measures are proposed. (To be presented in Spanish.)