Fri · 23 October · 15:00 → Fri · 23 October · Check event page for more details.
Get up-to-date information at: https://luma.com/backchannel-berlin Address: Check event page for more details. A research discussion on building frontier AI models, for people working on speech, audio and language models. End-to-end speech-to-speech models can now listen while they talk, handle interruptions and keep a consistent voice. Getting them to frontier quality is still largely open research: data, architecture, training objectives and evaluation. Running them for consumer apps, in every language and within a few hundred milliseconds, adds constraints of its own. The afternoon covers both. Open questions: speech-to-speech and conversation What should a full-duplex model predict, and at what… Hosted by Kylan Gibbs & 4 others