David Rossiter

Voice development under training with and without the influence of real‐time visually presented biofeedback

This paper describes an investigation into the developmental nature of the voice under training with and without the influence of real-time visually presented biofeedback. Two subjects who had not previously experienced any form of vocal training took six singing lessons. One was taught conventionally, while the other was taught with the aid of a system known as Acoustic and Laryngeal Biofeedback Enhancement Real Time (ALBERT). Real-time biofeedback was presented based upon measures of (i) larynx closed quotient (CQ), (ii) spectral amplitude in the singers formant frequency band relative to the spectral amplitude of the full band (ratio), and (iii) both parameters combined in a manner based on previously observed correlations between them. Results indicate generally increased sound pressure levels (SPL) of acoustic output and generally consistent increases in the level of CQ and ratio across consecutive lessons for both subjects.

Streaming Multimedia Delivery in Web Services Based E-Learning Platforms

The use of streaming multimedia has become more and more common as the content of e-learning material. The distribution of the material is typically achieved by a dedicated streaming server. There are several proprietary technologies that can act as the streaming server but there is a lack of standardized approach for this purpose. Recently importance has been placed on the interoperability and reusability of learning material by making use of standards such as SCORM and web services. In this paper we propose a service oriented architecture for the delivery of multimedia content which is interoperable and reusable. We model this delivery by extending current web service standards, and we briefly discuss the use of such web service resources in SCORM.

A SOAP-Based Streaming Content Delivery Framework for Multimedia Web Services

Web services provide efficient development and deployment of interoperable and reusable applications. Multimedia Web services is an implementation of multimedia applications within the Web services framework. However, current Web services standards are not sufficient to cater for multimedia streaming, the commonly used delivery method for large multimedia objects. In this paper, a SOAP-based streaming content delivery framework for multimedia Web services is proposed. The framework uses an extension of existing SOAP standards which allows streaming content to be transmitted between two SOAP nodes over the HTTP protocol. The extension includes a new SOAP streaming message exchange pattern and its corresponding SOAP HTTP binding. The proposed framework is capable of delivering quality adaptive multimedia content to users with heterogeneous configurations and requirements. By incorporating a proxy server between the service provider and requester the efficiency of the content delivery can be improved using multimedia caching.

Observed change in mean speaking voice fundamental frequency of two subjects undergoing voice training

Previous studies have indicated that trained and untrained voice users exhibit a mean speaking voice fundamental frequency (FO) at a particular percentage point relative to their complete voice FO range. However, as these observations are based on separate groups of subjects, it may be that there are physiological reasons for this. This paper presents a pilot investigation into the relationship between FO and complete voice range. For the study an assessment is made of the mean FO performance of two subjects assessed before, during and after a short course of vocal tuition. It is observed that the mean speaking voice fundamental frequency of both subjects exhibits a generally consistent tendency towards a point at approximately 17% of the total FO range.

The gong system: web-based learning for multiple languages, with special support for the Yale representation of Cantonese

This paper introduces the Gong system, an Internet-based voice board system designed primarily for language learners which includes special support for Cantonese. The Gong system is a client/server design which may be used to complement or, in some contexts, to replace face to face learning. The system supports Unicode input, storage and display of multiple character sets. Furthermore, we have developed a unique storage and display method for the Yale romanized representation of Cantonese, which is the most popular written method used for teaching Cantonese.

An investigation into the modelling of virtual objects with sound vibration properties

In order to increase the power of virtual environments, several different attempts have been made to incorporate sound interactivity in some form. For example, several implementations of virtual environments permit the playing of a previously recorded soundfile upon the triggering of an associated event. The user may then, for instance, perceive the sound of a creaky door when one is opened. However, a relatively more effective system for entertaining joint audio and visual response may be derived by using physical modelling techniques. We have undertaken a pilot investigation in which virtual objects are implemented in a manner such that they implicitly possess vibration properties analogous to that of the real world. Consequently these objects are able to vibrate in response to stimulus. The vibrations may be visually perceived as, for example, wave patterns on the surface of an object, and acoustically perceived by mapping values representative of surface displacement to a loudspeaker. This paper discusses the current state of the project.