Information theoretic acoustic feature selection for acoustic-to-articulatory inversion
نویسندگان
چکیده
We use mutual information as the criterion to rank the Mel frequency cepstral coefficients (MFCCs) and their derivatives according to the information they provide about different articulatory features in acoustic-to-articulatory (AtoA) inversion. It is found that just a small subset of the coefficients encodes maximal information about articulatory features and interestingly, this subset is articulatory feature specific. We use these subsets of MFCCs(+derivatives) in AtoA inversion using Gaussian mixture model (GMM) mapping. Inversion experiments with articulatory data support the information theoretic finding that the subsets of MFCCs(+derivatives) as selected by feature ranking method are sufficient to achieve an inversion performance similar to that obtained by a conventional full set of MFCCs(+derivatives). This drastically reduces the modeling complexity of the acoustic-articulatory map using GMM without degrading inversion performance significantly.
منابع مشابه
Information theoretic analysis of direct and estimated articulatory features for phonetic discrimination
It is well known that machine recognition of speech can be improved by including direct articulatory evidence in addition to the signal information derived from the acoustic speech. This has been shown through automatic phonetic recognition experiments [1] as well as by information theoretic analysis between phonetic classes and the acoustic/articulatory features [2]. However, access to such di...
متن کاملPalate-referenced articulatory features for acoustic-to-articulator inversion
The selection of effective articulatory features is an important component of tasks such as acoustic-to-articulator inversion and articulatory synthesis. Although it is common to use direct articulatory sensor measurements as feature variables, this approach fails to incorporate important physiological information such as palate height and shape and thus is not as representative of vocal tract ...
متن کاملPronunciation analysis by acoustic-to-articulatory feature inversion
Second language learners may require assistance correcting their articulation of unfamiliar phonemes in order to reach the target pronunciation. If, e.g., a talking head is to provide the learner with feedback on how to change the articulation, a required first step is to be able to analyze the learner’s articulation. This paper describes how a specialized restricted acoustic-to-articulatory in...
متن کاملAudiovisual-to-articulatory inversion
It has been shown that acoustic-to-articulatory inversion, i.e. estimation of the articulatory configuration from the corresponding acoustic signal, can be greatly improved by adding visual features extracted from the speaker’s face. In order to make the inversion method usable in a realistic application, these features should be possible to obtain from a monocular frontal face video, where the...
متن کاملSpeaker verification based on fusion of acoustic and articulatory information
We propose a practical, feature-level fusion approach for speaker verification using information from both acoustic and articulatory signals. We find that concatenating articulation features obtained from actual speech production data with conventional Mel-frequency cepstral coefficients (MFCCs) improves the overall speaker verification performance. However, since access to actual speech produc...
متن کامل