Search

Freedom Fone (FF) is an Interactive Voice Response (IVR) System that integrates with the Global System for Mobile (GSM) telecommunications [1]. Sahana is a disaster information management expert system working with Internet technologies [2]. The Project intent was to mediate information between the FF and Sahana through the Emergency Data Exchange Language (EDXL) interoperable content standard [3]. It goal was to equip Sarvodaya, Sri Lanka's largest humanitarian organization, with voice-enabled disaster communication. The 3.52 Mean Opinion Score (MOS) for voice quality was an early automation challenge in introducing Automatic Speech Recognition (ASR). A 4.0 MOS was determined as a cut-point for classifying reliable voice data [4]. The Percent Difficult (PD) in an emulated speaker-independent scenario was 29.44% and a speaker-dependent scenario was 13.24%. Replacing human operators with ASR software proved inefficient [5] and [6]. This paper discusses uncertainties that are barriers to integrating voice enabled automated emergency communication services for response resource analysis and decision support. [ABSTRACT FROM PUBLISHER]

This paper focuses on the automatic modification of the degree of articulation (hypo/hyperarticulation) of an existing standard neutral voice in the framework of HMM-based speech synthesis. Starting from a source speaker for which neutral, hypo and hyperarticulated speech data are available, two sets of transformations are computed during the adaptation of the neutral speech synthesizer. These transformations are then applied to a new target speaker for which no hypo/hyperarticulated recordings are available. Four statistical methods are investigated, differing in the speaking style adaptation technique (MLLR vs. CMLLR) and in the speaking style transposition approach (phonetic vs. acoustic correspondence) they use. This study focuses on the prosody model although such techniques can be applied to any stream of parameters exhibiting suited interpolability properties. Two subjective evaluations are performed in order to determine which statistical transformation method achieves the better segmental quality and reproduction of the articulation degree. [ABSTRACT FROM PUBLISHER]

This paper introduces an entropy-based spectral representation as a measure of the degree of noisiness in audio signals, complementary to the standard MFCCs for audio and speech recognition. The proposed representation is based on the Rényi entropy, which is a generalization of the Shannon entropy. In audio signal representation, Rényi entropy presents the advantage of focusing either on the harmonic content (prominent amplitude within a distribution) or on the noise content (equal distribution of amplitudes). The proposed representation outperforms all other noisiness measures — including Shannon and Wiener entropies — in a large-scale classification of vocal effort (whispered-soft/normal/loud-shouted) in the real scenario of multi-language massive role-playing video games. The improvement is around 10% in relative error reduction, and is particularly significant for the recognition of noisy speech — i.e., whispery/breathy speech. This confirms the role of noisiness for speech recognition, and will further be extended to the classification of voice quality for the design of an automatic voice casting system in video games. [ABSTRACT FROM PUBLISHER]

Audiobooks are known to contain a variety of expressive speaking styles that occur as a result of the narrator mimicking a character in a story, or expressing affect. An accurate modeling of this variety is essential for the purposes of speech synthesis from an audiobook. Voice quality differences are important features characterizing these different speaking styles, which are realized on a gradient and are often difficult to predict from the text. The present study uses a parameter characterizing breathy to tense voice qualities using features of the wavelet transform, and a measure for identifying creaky segments in an utterance. Based on these features, a combination of supervised and unsupervised classification is used to detect the regions in an audiobook, where the speaker changes his regular voice quality to a particular voice style. The target voice style candidates are selected based on the agreement of the supervised classifier ensemble output, and evaluated in a listening test. [ABSTRACT FROM PUBLISHER]

This paper presents a method for reducing a bumblebee noise generated by a GSM technology in a smartphone. Global smartphone penetration has been very swift and 2nd generation, 3rd generation and 4th generation communication technology are commercially used in the world. The 2nd generation wireless telephone technology is based on two standards, depending on the type of multiplexing used the Global System for Mobile Communication (GSM) and Code-Division Multiple Access (CDMA). In particular, the GSM technology has a market share of almost more than 79% in the world. GSM technology uses a channel access method that combines frequency division multiple access (FDMA) and time division multiple access (TDMA). There are four commercial frequency bands. GSM technology has a burst structure by a TDMA method. And hence, the GSM technology has a disadvantage; radiation noise is generated from an antenna propagation signal of the smartphone, and consequently, the voice quality of the smartphone is degraded. This noise is commonly known as bumblebee noise, buzz noise or TDMA noise. There have been several studies to reduce the noise since a release of GSM technology in a commercial market. Those studies mainly focused on designing infinite impulse response (IIR) notch filters by the signal processing technology or on data burst transmission schemes. The first method needs more million instructions per second (MIPS) for operation, and hence, current consumption increases accordingly. The latter method requires a change in the protocol standard of the GSM; hence, its feasibility is low. This paper presents hardware filters designed to reduce the bumblebee noise in smartphones, especially the bumblebee noise in the headset mode during calls. The bumblebee noise is easily generated in the headset mode of smartphones during calls because the headset is necessarily connected to the smartphone. This paper verified that, on an average the bumblebee noise was reduced up to 14dBm by experiments. [ABSTRACT FROM PUBLISHER]

Prior research on voice has identified important managerial traits and behaviors, such as openness and solicitation, and voice climate conditions that influence frequency of employee voice. This collection of symposium papers helps deepen our understanding of how managers and the team context influence the voice process by adding new team and manager-related antecedents into the equation, such as group norms, participative leadership, leader personal control, leader framing of voice significance, and leader rejection tactics. In addition, several of these papers add complexity to our scholarly thinking around voice by exploring how these manager-related variables interact with other contextual variables to more precisely describe effects on voice behavior. Lastly, by examining antecedents of voice quality (versus voice frequency, which has been the focus of most prior voice research), this symposium provides some practical guidance to managers who want to encourage not only more employee voice, but better employee voice, which in turn should enhance voice benefits to the organization. [ABSTRACT FROM AUTHOR]