Articulatory Effort in Different Speaking Rates
نویسندگان
چکیده
Articulatory effort has been widely discussed under different speech production models. A good way to explore this issue is to test the hypothesis that speakers naturally desire to minimize articulatory effort during fast speech rates, such as undershoot of the target sound. This study demonstrates articulatory effort at different speaking rates by examining articulatory trajectory using the Electromagnetic Articulograph AG500. The results suggest that the articulator undershoots a target and the valley of the target might go deeper while the velocity increases.
منابع مشابه
Manipulation of the prosodic features of vocal tract length, nasality and articulatory precision using articulatory synthesis
Vocal emotions, as well as different speaking styles and speaker traits are characterized by a complex interplay of multiple prosodic features. Natural sounding speech synthesis with the ability to control such paralinguistic aspects requires the manipulation of the corresponding prosodic features. With traditional concatenative speech synthesis it is easy to manipulate the “primary” prosodic f...
متن کاملCompensating hyperarticulation for automatic speech recognition
This thesis details the effects of hyperarticulation in the context of automatic speech recognition used for human-to-machine interaction. Hyperarticulation can be characterised as a speaking mode exhibiting an exaggerated articulation and occurs as a natural reaction in an effort to resolve recognition errors. Despite the user’s attempt to disambiguate word confusions, hyperarticulation causes...
متن کاملArticulatory features for "meeting" speech recognition
“Meeting” speech, for example from the RT-04S task, contains a mixture of different speaking styles that leads to word error rates higher than 25% even when close-talking microphones are being used. The problem is even more serious, as word error rates are particularly high when speakers use a clear speaking mode, for example because they want to stress an important point. Previous work showed ...
متن کاملCross-language Differences in Overlap and Assimilation Patterns in Korean and Russian
This paper investigates cross-linguistic differences in gestural overlap in consonant clusters and discusses how different patterns of overlap may interact with language-specific place assimilation patterns. We examine Russian and Korean stopstop sequences within and across words, produced at two speaking rates. Significant differences in degrees of overlap emerge between the two languages for ...
متن کاملCompensating for Hyperartic Articulatory Pro
In spoken dialogue systems, hyperarticulation occur as an effect to recover previous recognition errors. It is commonly observed that users of automatic speech recognition systems apply similar recovery strategies as in human-human interactions. Previous studies have shown that current speech recognizers don’t cover hyperarticulated speech well. As an effect of higher word error rates at hypera...
متن کامل