PhD Studentship on 'Communicative/Expressive Speech Synthesis' at UNIVERSITY of SHEFFIELD
Recent years have seen a substantial growth in the capabilities of Speech Technology
systems, both in the research laboratory and in the commercial marketplace.
However, despite this progress, contemporary speech technology is not able to
fulfil the requirements demanded by many potential applications, and performance
is still significantly short of the capabilities exhibited by human talkers
and listeners, especially in interactive real-world environments.
This shortfall is especially noticeable in the 'text-to-speech' (TTS) systems
that have been developed for automated spoken language output.
Considerable advances have been made in naturalness and voice quality, yet state-of-the-art
TTS systems still exhibit a rather limited range of speaking styles, a general
lack of expressiveness and restricted communicative functionality.
The objective of this research is to investigate novel approaches to text-to-speech synthesis that have the potential to overcome these limitations, and which could contribute to the next-generation of speech-based systems, especially in application areas such as assistive technology.
Funding is available immediately for an eligible UK/EU student.
Applicants should possess a computational background and should ideally have
some knowledge/experience of speech processing.
Thesis Supervisor: Prof. Roger K. Moore
For further information, contact Prof. Roger Moore or see our website for how to apply.
The Speech and Hearing research group in Computer Science at the University of Sheffield has an international reputation in the multi-disciplinary field of speech and hearing research. With three chairs, four faculty, five research associates and around twelve research students, this is one of the strongest teams worldwide. A unique aspect of the group is the wide spectrum of research topics covered, from the psychophysics of hearing through to the engineering of state-of-the-art speech technology systems.