Distance score evaluation of the visualised speech spectra at audio-visual articulation training
نویسندگان
چکیده
In the frame of the Inco-Copernicus program of the European Commission titled „A Multimedia Multilingual Teaching and Training System for Speech Handicapped children” an audiovisual pronunciation teaching and training method and software system has been developed for hearing and speechhandicapped persons to help them to control their speech production. During a part of the training the interpretation of the signal is based on the comparisons of the signal with the stored references. The aim of the present study is to find a distance measure that can help these comparisons and mirror the judgement of the listeners. Three spectral distance calculations have been compared. The good and unacceptable examples were separated well on the base of the Average Spectrum Distance calculation. This calculation can be the base of an automatic feedback of the actual pronunciation that could approach the decision of the listeners well.
منابع مشابه
Innovations in Czech audio-visual speech synthesis for precise articulation
This paper presents new steps toward animation of precise articulation. The acquisition of audio-visual corpus for Czech and new method for parameterization of visual speech was designed to obtain exact speech data. The parameterization method is primarily suitable for training a data driven visual speech synthesis systems. The audio-visual corpus includes also specially designed test part. Fur...
متن کاملComparison of Motor Skills Among Studens with Intellectual Disability, Stuttering, Articulation Problems and Normal Speech
Objective: This research aimed to compare the motor skills among students with intellectual disability, stuttering, articulation problems and normal speech. Methods: The study was a retrospective causal-comparative research. From among all elementary male students with intellectual disability in Urmia city, 90 students (30 students in each group) were selected. All groups completed the revised ...
متن کاملAn Expandable W Audiovisual Text-to-Speech
The authors propose a framework for audiovisual speech synthesis systems [1] and present a first implementation of the framework [2], which is called MASSY Modular Audiovisual Speech SYnthesizer. This paper describes how the audiovisual speech synthesis system, the ‘talking head’, works, how it can be integrated into web-applications, and why it is worthwhile using it. The presented application...
متن کاملSpeaker normalization for audio-visual articulation training
The paper describes formant based speaker normalization method suitable for speech visualization and articulation training systems. The method estimates the error function obtained from speaker formant characteristics for a given vowel. Estimated error function gives information for critical band filter shifting on mel-warped frequency scale. The paper also describes accurate technique for form...
متن کاملMASSY - a Prototypic Implementation of the Modular Audiovisual Speech SYnthesizer
Audiovisual speech synthesis systems usually are inflexible with respect to the ability to replace the audio and video synthesis and the control algorithms due to the dependencies of the implemented pieces. In order to enable a newly developed system to exchange modules, to evaluate their specific advantages, and to detect their weak points, the author proposes a framework for audiovisual speec...
متن کامل