This work proposes a novel method of predicting formant frequencies from a stream of mel-frequency cepstral coefficients (MFCC) feature vectors. Prediction is based on modelling the joint density of MFCCs and formant frequencies using a Gaussian mixture model (GMM). Using this GMM and an input MFCC vector, two maximum a posteriori (MAP) prediction methods are developed. The first method predict...

This paper describes how formant frequencies of voiced and unvoiced speech can be predicted from mel-frequency cepstral coefficients (MFCC) vectors using maximum a posteriori (MAP) estimation within a hidden Markov model (HMM) framework. Gaussian mixture models (GMMs) are used to model the local joint density of MFCCs and formant frequencies. More localised prediction is achieved by modelling s...

this paper reports the results of an experimental study on non-native production of english vowels. two groups of persian efl learners varying in language proficiency were tested on their ability to produce the nine plain vowels of american english. vowel production accuracy was assessed by means of acoustic measurements. ladefoged and maddison’s (1996) f1 f2 measurements for american english v...

Acoustical analysis of speech and perceptual studies indicate that the dominant acoustic correlates of vowel perception are the frequencies of the first three formants. However, most vowels are not completely steady-state (even in isolation) and formant frequencies change with variation in the surrounding consonantal context, prosodic influences, speaking rate, and vocal tract length of the tal...

In the present study Dutch vowel formant characteristics of laryngectomized tracbeoesophageal speakers are investigated.. Both vowels in CV nonsense syllables (/al, Iii, and /u/), and vowels of stressed syllables in read-aloud text are studied. It appeared that in the nonsense syllables the first formants were comparable to those ·of normal speakers, but that the second formants were higher tha...

This paper describes a new method for estimating formant frequencies. It operates in two phases. The first phase, which is similar to a technique developed by Talkin [JASA, vol. 82, S1], finds optimal formant track estimates by imposing frequency continuity constraints using Dynamic Programming (DP). DP is used to select a mapping of candidate frequencies to formant frequencies in oral sonorant...

Spectra of vowels sung a t identical pitches by two tenor voices and two alto voices a r e compared. The formant frequencies and the source spectra a r e studied by matching the spectra on a terminal analogue of the vocal tract. As regards the source spectrum the alto voices show a higher relative amplitude of the fundamental than the tenor voices. As regards the formants, ari apparent differen...