No abstract

No abstract

Stuttering is a speech disorder during which the flow of speech is interrupted by involuntary pauses and repetition of sounds. Stuttering identification is an interesting interdisciplinary domain research problem which involves pathology, psychology, acoustics, and signal processing that makes it hard and complicated to detect. Recent developments in machine and deep learning have dramatically revolutionized speech domain, however minimal attention has been given to stuttering identification. This work fills the gap by trying to bring researchers together from interdisciplinary fields. In this paper, we review comprehensively acoustic features, statistical and deep learning based stuttering/disfluency classification methods. We also present several challenges and possible future directions.

De l'utilisation de la pause silencieuse dans le débat politique télévisé. Le cas de François Hollande « Si l'oeil reste rieur, la voix se casse au fil des meetings. […] Alors son débit à la tribune se densifie. Les mots sont ramassés, entrecoupés de longs silences puis à nouveau une rafale explosive, une véritable compression rythmique, la voix saturée prend des harmoniques aigus. » (Le Monde, 4 mai 2012) Ces remarques au sujet de François Hollande, lues dans une tribune publiée par André Manoukian sur le site du Monde, sont révélatrices du fait que la forme du discours est tout aussi importante que le contenu, dans la mesure où la première participe pleinement à la construction d'une image de présidentiable. En outre, ces mêmes phrases sont la preuve de la singularité de l'élocution propre à François Hollande, qui est notamment définie ici par un débit rapide et de longues interruptions. L'objectif de cet article est justement d'étudier le rythme dans le discours de François Hollande, en se focalisant plus particulièrement sur l'utilisation qu'il fait des pauses, et ce afin de comprendre les spécificités de sa parole. Notre hypothèse est que le placement des pauses est loin d'être anodin, dans la mesure où elles contribuent à la structuration et à la mise en lumière de certaines parties du discours. Ces interruptions rythmiques seront étudiées à travers deux débats politiques télévisés en face-à-face.

Stuttering is a neuro-developmental speech impairment characterized by uncontrolled utterances (interjections) and core behaviors (blocks, repetitions, and prolongations), and is caused by the failure of speech sensorimotors. Due to its complex nature, stuttering detection (SD) is a difficult task. If detected at an early stage, it could facilitate speech therapists to observe and rectify the speech patterns of persons who stutter (PWS). The stuttered speech of PWS is usually available in limited amounts and is highly imbalanced. To this end, we address the class imbalance problem in the SD domain via a multibranching (MB) scheme and by weighting the contribution of classes in the overall loss function, resulting in a huge improvement in stuttering classes on the SEP-28k dataset over the baseline (StutterNet). To tackle data scarcity, we investigate the effectiveness of data augmentation on top of a multi-branched training scheme. The augmented training outperforms the MB StutterNet (clean) by a relative margin of 4.18% in macro F1-score (F 1 ). In addition, we propose a multi-contextual (MC) StutterNet, which exploits different contexts of the stuttered speech, resulting in an overall improvement of 4.48% in F 1 over the single context based MB StutterNet. Finally, we have shown that applying data augmentation in the cross-corpora scenario can improve the overall SD performance by a relative margin of 13.23% in F 1 over the clean training.