Interspeech 2022
Three members of the team travelled to Incheon, South Korea in September for Interspeech 2022 where we chaired and presented at the following sessions:
- Speech Production on Monday the 19th: Exploration strategies for articulatory synthesis of complex syllable onsets [slides, presentation, samples]
- Speech Synthesis: New Applications on Tuesday the 20th: Articulatory Synthesis for Data Augmentation in Phoneme Recognition
- Show and Tell III on Wednesday the 21st: Evoc-Learn — High quality simulation of early vocal learning [poster]
Project Workshop in Macedonia
In July 2022, Branislav Gerazov hosted a Workshop at Lake Ohrid in Macedonia where the team presented and reflected on aspects of the project.
The following talks were presented:
- Yi Xu — High Quality Simulation of Early Vocal Learning [presentation]
- Santitham Prom-on — Learning of Thai Syllables [presentation]
- Peter Birkholz — VocalTractLab Features and Roadmap [presentation]
- Daniel van Niekerk — Learning CVs and CCVs [presentation]
- Anqi Xu — A computational simulation of human vocal learning [presentation]
- Paul K. Krug — Artificial Vocal Learning guided by Phoneme Recognition and Visual Information [presentation]
- Branislav Gerazov — ASR for early vocal learning [presentation]
Interspeech 2021
Anqi Xu presented an analysis of the relationship between articulatory and acoustic spaces for British vowels at Interspeech 2021:
- Model-based exploration of linking between vowel articulatory space and acoustic space [samples]
Interspeech 2020
Daniel van Niekerk presented a paper at Interspeech 2020 testing the use of automatic speech recognition (ASR) as an evaluator of articulatory-produced CV syllables discovered using different optimisation algorithms:
- Finding intelligible consonant-vowel sounds using high-quality articulatory synthesis [slides, presentation]
