Voice-Prosody and Technologies to Give Voice to Endangered Languages: an Irish Perspective

Add to your list(s) Download to your calendar using vCal

• Ailbhe Ní Chasaide (Trinity College Dublin)
• Thursday 26 November 2020, 16:30-18:00
• Online.

If you have a question about this talk, please contact Tim Laméris.

Please register by noon on the day of the talk: https://forms.gle/pW2NTo3WHdVe59vD6

Registration link https://forms.gle/pW2NTo3WHdVe59vD6

ABSTRACT

This paper presents research conducted along with researchers at our laboratory, looking at the voice, human and synthetic, from three rather different, but complementary perspectives. Much of this work has focussed on Irish. The first is the study of intonation, or melodic modulation in the Irish dialects. Striking differences emerge between Northern and Southern dialects. Even within a single dialect region some striking sub-dialect signatures emerge, in the time-alignment of melody and syllables. The second area concerns our generally more holistic view of prosody, as being not just about melodic modulation, but rather a gestalt to which modulation of voice quality (tone-of-voice) features importantly contributes. It is argued that a full account and understanding of intonation will require that we encompass all dimensions of the voice – bearing in mind of course that capturing voice source modulation is a challenging task. Such a holistic approach would also open up the area of affect in prosody, a central aspect of prosody that is largely neglected within linguistics. The third area I discuss concerns the building of synthetic voices for the Irish dialects, which has been ongoing in the ABAIR project (www.abair.ie). This, and the derivative technologies that have followed, are giving voice to – and supporting – the language community, in often unforeseen ways. The impact of this work resonates well beyond academia and this research offers the linguist (in collaboration with technologists) ways, not only to document a language, but also to have a real impact on its survival. Going forward, we are currently exploring how our insights on voice prosody can be implemented in our synthetic voices. This provides a test bed for voice research and opens the door to future (synthetic) voice-based interactive systems, which will hopefully capture some of the diversity and affective dimensions of human prosody.

This talk is part of the Cambridge Linguistics Forum series.

This talk is included in these lists:

• Cambridge Forum of Science and Humanities
• Cambridge Language Sciences
• Cambridge Linguistics Forum
• Cambridge talks
• Chris Davis' list
• Featured lists
• Guy Emerson's list
• Language Sciences for Graduate Students
• Online

Note that ex-directory lists are not shown.