Whispering is a natural, unphonated, secondary aspect of speech communications for most people. However, it is the primary mechanism of communications for some speakers who have impaired voice production mechanisms, such as partial laryngectomees, as well as for those prescribed voice rest, which often follows surgery or damage to the larynx. Unlike most people, who choose when to whisper and when not to, these speakers may have little choice but to rely on whispers for much of their daily vocal interaction.Even though most speakers will whisper at times, and some speakers can only whisper, the majority of today’s computational speech technology systems assume or require phonated speech. This article considers conversion of whispers into natural-sounding phonated speech as a noninvasive prosthetic aid for people with voice impairments who can only whisper. As a by-product, the technique is also useful for unimpaired speakers who choose to whisper.Speech reconstruction systems can be classified into those requiring training and those that do not. Among the latter, a recent parametric reconstruction framework is explored and then enhanced through a refined estimation of plausible pitch from weighted formant differences. The improved reconstruction framework, with proposed formant-derived artificial pitch modulation, is validated through subjective and objective comparison tests alongside state-of-the-art alternatives.

/pdf/reconstruction-of-phonated-speech-from-whispers-using-539rf82alz.pdf

Reconstruction of Phonated Speech from Whispers Using Formant-Derived Plausible Pitch Modulation

In this work, we address the issues involved in whisper-to-audible speech conversion. Spectral mapping techniques using Gaussian mixture models or Artificial Neural Networks borrowed from voice conversion have been applied to transform whisper spectral features to normally phonated audible speech. However, the modeling and generation of fundamental frequency ($F_0$) and its contour in the converted speech is a major issue. Whispered speech does not contain explicit voicing characteristics and hence it is hard to derive a suitable $F_0$, making it difficult to generate a natural prosody after conversion. Our work addresses the $F_0$ modeling in whisper-to-speech conversion. We show that $F_0$ contours can be derived from the mapped spectral vectors, which can be used for the synthesis of a speech signal. We also present a hybrid unit selection approach for whisper-to-speech conversion. Unit selection is performed on the spectral vectors, where $F_0$ and its contour can be obtained as a byproduct without any additional modeling.

/pdf/fundamental-frequency-generation-for-whisper-to-audible-y29slsx0qg.pdf

Fundamental frequency generation for whisper-to-audible speech conversion

For voice handicapped people, an easy to use voicing aid device is wanted. In mobile telephony, so-called non-speaking speech communication is an expected solution for essential privacy as well as for acoustic nuisance prevention. The study introduced here intends to cover both issues, introducing a system where the whispering (non-speaking voice or talk without vocal fold activation) signal is converted to pseudo-real voice signal, which is to be sent to, or heard by, the other party. The study also includes validation tests with multiple volunteers for its output legibility. Unlike general concept of speech regeneration being inclined to signal recognition or decomposition to text followed by electronic reading (voicing), our system converts it almost directly without recognition or decomposition steps. The processing is based on repetitive playback of short time autocorrelation, conducted by synthetic pitch pulse. A real time software pipeline process is under development.

Transformation of whispering voice to pseudo-real voice for unvoiced telephony and communication aid for voice-handicapped persons

Reconstruction of whispering voice into a pseudo-real voice was previously covered by other researchers using statistical processes. We've been doing this conversion from a different perspective and using the raw signal data only. As a result we can avoid heavy computation load because our approach does not require the data to be transformed before being analyzed. After some good results obtained in laboratory we've questioned ourselves about the quality of the reconstructed samples to real people and the results are presented in this study.

http://ieeexplore.ieee.org/iel5/6088936/6100662/06100741.pdf

An analysis of human acceptability of recreated voice samples from whispering

Abstract A possible problem of orthogonal frequency division multiplexing (OFDM) is its high peak-to-average power ratio (PAPR). Recently, the single carrier (SC) transmission system with frequency-domain equalization is attracting much attention. In this paper, a novel approach to alleviate the PAPR problem of OFDM is proposed. A generalized OFDM (GOFDM) with frequency-domain equalization is presented and its performance in a frequency-selective fading channel is evaluated and compared with those of conventional OFDM and single carrier (SC) systems.

Generalized OFDM with Frequency-domain Equalization (移動通信ワークショップ)

Analyzing the whispering voice signal, it is considered to be as understandable as normal speech when heard at proximity and can also carry prosodic information. The study introduced here is centered on developing a signal processing application, not a signal recognizing application, where the whispering (non-speaking voice or talk without vocal fold activation) signal is converted to pseudo-real voice signal similar to normal speech. The method proposed in this study tries to achieve its objective without relying on any specific hardware device, making use of the whispering voice signal's own properties and information. Finally, validation of the proposed method has been conduced and the results analyzed.

Anderson Pierre Passos

Papers

Transformation of whispering voice to pseudo-real voice for unvoiced telephony and communication aid for voice-handicapped persons

A lightweight processing for conversion of whispering voice into normal speech