Extraction and Utilization of Excitation Information of Speech: A Review

Abstract

| Speech production can be regarded as a process where a time-varying vocal tract system (filter) is excited by a time-varying excitation. In addition to its linguistic message, the speech signal also carries information about, for example, the gender and age of the speaker. Moreover, the speech signal includes acoustical cues about several speaker traits, such as the emotional state and the state of health of the speaker. In order to understand the production of these acoustical cues by the human speech production mechanism and utilize this information in speech technology, it is necessary to extract features describing both the excitation and the filter of the human speech production mechanism. While the methods to estimate and parameterize the vocal tract system are well established, the excitation appears less studied. This article provides a review of signal processing approaches used for the extraction of excitation information from speech. This article highlights the importance of excitation information in the analysis and classification of phonation type and vocal emotions, in the analysis of nonverbal laughter sounds, and in studying pathological voices. Furthermore, recent developments of deep learning techniques in the context of extraction and utilization of the excitation information are discussed.

Measuring pitch extractors' response to frequency-modulated multi-component signals

“…The assumed signal model for this measurement is a variant of the sinusoidal model (McAulay and Quatieri, 1986). We need to explore pitch extractors assuming the other signal models, especially excitation-based models (Kadiri et al, 2021). We also need to explore relations to deep-learning-based pitch extractors.…”

Enhancement in Speaker Identification through Feature Fusion using Advanced Dilated Convolution Neural Network

“…However, it is necessary to understand the acoustic cues that describe the detailed speaker characteristics, including different voice qualities and emotions. Then, utilize this information to identify the speaker [12]. Hence, it is required to capture the features describing excitation and the vocal filter in speech production.…”

“…Thus, excitation features provide supportive information to the frequently used vocal tract features of various speakers. The different methods are well established to describe the vocal tract filter, but the researchers showed less interest in excitation features [12]. The study of [13] demonstrated the methods to capture the excitation features effectively and mentioned the future scope to combine excitation and vocal tract features.…”

“…They picked up the log-MelSpectrum of the speech signal and obtained a notable improvement in accuracy. In [12], the authors stated that it is required to analyze the generation of acoustical cues in the process of speech production. Extracting the features of both the excitation and vocal tract system is required.…”

“…It includes the emotion and health state of the speaker. The speech production system consists of two functional parts [12]. One is excitation produced at the larynx and another is filtering, which is excited by the excitation e(n).…”