Formant extraction from group delay function
نویسندگان
چکیده
This paper presents an approach based on the properties of group delay functions for extracting formants from speech signals. The algorithm is similar to the cepstral smoothing approach for formant extraction using homomorphic deconvolution. The significant differences are (i) the logarithmic operation is replaced by ()' operation and (ii) the additive and high resolution properties of group delay functions are exploited to emphasize formant peaks. The group delay function (or the negative derivative of the Fourier transform phase) is derived for a signal which in turn is derived from the Fourier transform magnitude of the speech signal. If a suitable value of r is used, this method gives highly consistent estimates of formants compared to both the cepstral approach and the model-based linear prediction (LP) approach for smoothing the magnitude spectrum. The effects of the parameters, exponent r and window width p, on the proposed technique for formant extraction are studied. Zusammenfassung. Dieser Beitrag stellt eine Methode zur Messung der Formantfrequenzen vor welche die Eigenschaften der Gruppenlaufzeiffunktionen ausniitzt. Der Algorithmus ist der kepstralen Methode zur spektralen Abrundung ~hnlich. Die zwei wichtigsten Underschiede sind (1) der Logarithmus wird durch einen ()r Operator ersetzt und (2) die additiven Eigenschaften und das gute AuflEsungsverm6gen der Gruppenlaufzeitfunktionen werden ausgenutzt um die Scheitelpunkte der Formanten hervorzuheben. Die Gruppenlaufzeitfunktionen (oder die negative Ableitung der Phase des Fourierspektrums) wird abgeleitet fiJr ein Signal welches seinerseits von der Magnitude des Fourierspektrums des Sprachsignals abgeleitet wird. Wenn ein passender Wert fiir r gebraucht wird, dann ergibt die Methode Sch~itzwerte ffir die Formanten welche vergleichbar sind mit denen welche mit der kepstralen Methode oder mit der linearen Pr~idiktion gewonnen werden. Di¢ Auswirkung des Exponenten r sowie der L~inge des Analysefensters auf die Ergebnisse werden ebenfalls untersucht. Risum6, Ce papier prEsente une technique fondEe sur les propriEtEs des fonctions retard de groupe afin d'extraire les formants des signaux de parole. L'algorithme est semblable au lissage cepstral utilisant la dEconvolution homomorphique. Les differences significatives sont les suivantes: (a) le logarithme est remplacE par un opErateur ()r et (b) les propriEtEs additive et de haute resolution des fonctions retard sont exploitEes pour accentuer les crates des formants. La fonction retard de groupe (ou la dErivEe negative de la phase de la transformEe de Fourier) est dErivEe pour un signal qui, ~ son tour, est dEriv6 de l'amplitude de la transformEe de Fourier du signal. Si une valeur convenable de rest utilisEe, cette mEthode donne des estimations formantiques tr~s cohErentes compar6es ~ eelles obtenues par la technique cepstrale ou par la prediction linEaire. Les effets de l'exposant r et de la largeur de la fenEtre sur la technique proposEe ont 6t6 6tudiEs.
منابع مشابه
The analysis on band-limited hypernasal speech using group delay based formant extraction technique
Speakers with defective velopharyngeal mechanism, produce speech with inappropriate nasal resonances across vowel sounds. The acoustic analysis on hypernasal speech and nasalized vowels of normal speech shows that there is an additional frequency introduced in the low frequency region close to the first formant frequency [1]. The conventional formant extraction techniques may fail to resolve cl...
متن کاملZeros of the z-transform (ZZT) representation and chirp group delay processing for the analysis of source and filter characteristics of speech signals
This study proposes a new spectral representation called the Zeros of Z-Transform (ZZT), which is an all-zero representation of the z-transform of the signal. In addition, new chirp group delay processing techniques are developed for analysis of resonances of a signal. The combination of the ZZT representation with the chirp group delay processing algorithms provides a useful domain to study re...
متن کاملRobust feature extraction based on spectral peaks of group delay and autocorrelation function and phase domain analysis
This paper presents a new robust feature set for noisy speech recognition in phase domain along with spectral peaks obtained from group delay and autocorrelation functions. The group delay domain is appropriate for formant tracking and autocorrelation domain is well-known for its pole preserving and noise separation properties. In this paper, we report on appending spectral peaks obtained in ei...
متن کاملModified Group Delay Based MultiPitch Estimation in Co-Channel Speech
Phase processing has been replaced by group delay processing for the extraction of source and system parameters from speech. Group delay functions are ill-behaved when the transfer function has zeros that are close to unit circle in the z-domain. The modified group delay function addresses this problem and has been successfully used for formant and monopitch estimation. In this paper, modified ...
متن کاملGroup-delay-deviation based spectral analysis of speech
In this paper, we investigate a new method for extracting useful information from the group delay spectrum of speech. The group delay spectrum is often poorly behaved and noisy. In the literature, various methods have been proposed to address this problem. However, to make the group delay a more tractable function, these methods have typically relied upon some modification of the underlying spe...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
- Speech Communication
دوره 10 شماره
صفحات -
تاریخ انتشار 1991