Analysis of a Modern Voice Morphing Approach using Gaussian Mixture Models for Laryngectomees

Abstract

This paper proposes a voicemorphing system for people suffering from Laryngectomy, which is the surgical removal of all or part of the larynx or the voice box, particularly performed in cases of laryngeal cancer. A primitive method of achieving voice morphing is by extracting the source's vocal coefficients and then converting them into the target speaker's vocal parameters. In this paper, we deploy Gaussian Mixture Models (GMM) for mapping the coefficients from source to destination. However, the use of the traditional/conventional GMM-based mapping approach results in the problem of over-smoothening of the converted voice. Thus, we hereby propose a unique method to perform efficient voice morphing and conversion based on GMM, which overcomes the traditional-method effects of over-smoothening. It uses a technique of glottal waveform separation and prediction of excitations and hence the result shows that not only oversmoothening is eliminated but also the transformed vocal tract parameters match with the target. Moreover, the synthesized speech thus obtained is found to be of a sufficiently high quality. Thus, voice morphing based on a unique GMM approach has been proposed and also critically evaluated based on various subjective and objective evaluation parameters. Further, an application of voice morphing for Laryngectomees which deploys this unique approach has been recommended by this paper.

Development of a hands‐free electrolarynx for obtaining a human‐like voice using the LPC residual wave

“…There are studies focusing on EL transducers (vibrators) to reduce specific noise components, [11][12][13][14] to reduce power consumption and improve quality of vowels, 15 to improve sound quality using machine learning, [16][17][18][19][20][21][22][23] and to realize voice conversion using both noise reduction and machine learning. 24 However, these methods use speakers to reproduce sound, which causes time delays in speech.…”

Voice morphing based on spectral features and prosodic modification

“…Another application is for the people suffering from Laryngectomy [11] -the surgical removal of the larynx (particularly performed in case of laryngeal cancer). In this case, extracting the source's vocal coefficients and converting them into the target speaker's vocal characteristics through training achieves voice morphing.…”