Details
- Starts
- Fri 23 Oct 2026, 13:00
- Ends
- Fri 23 Oct 2026, 18:00
- Type
- In person
- Category
- Tech & Startups
- City
- Berlin
- Country
- Germany
About this event
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, ha...
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
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
Links and contact
- Registration
- https://luma.com/backchannel-berlin
Source
- Imported from
- Luma
- Source city
- Berlin
- Original event
- Open source event https://luma.com/backchannel-berlin
