This session will look at why open source voice is so difficult, and what strategies are currently being applied to meet these challenges:
- Real time speech to text- including hardware limitations, difficulties with on-device STT, challenges with cloud based STT including responsiveness.
- Machine learning and training challenges - including debugging poor fit models, sourcing significant data sets and training new models.
- Multiple languages, localisation and internationalisation - including language idiosyncrasies, slang and handling intent collisions across multiple Skills
- Voice user interaction - the challenge of producing a voice interaction that is both useful AND human-like
Originally presented at
linux.conf.au
linux.conf.au is a conference about the Linux operating system, and all aspects of the thriving ecosystem of Free and Open Source Software that has grown up around it. Run since 1999, in a different Australian or New Zealand city each year, by a team of local volunteers, LCA invites more than 500 people to learn from the people who shape the future of Open Source.