JSALT 2025 Summer Workshop - Speech and Language Technology Research Opening Day

Experience the opening day of the prestigious JSALT 2025 summer workshop through this comprehensive 3.5-hour livestream recording from Brno University of Technology in the Czech Republic. Begin with welcoming remarks from university leadership and delve into the rich history of these influential summer workshops on speech and language technology presented by Sanjeev Khudanpur. Explore cutting-edge research presentations including EMMA's end-to-end multi-channel multi-talker automatic speech recognition system by Lukáš Burget, discover how large language models can maintain consistent role-playing behavior in Ricard Marxer's presentation on LLM agents, and learn about the ambitious AuGI project toward audio general intelligence from Ramani Duraiswami. Gain insights into text-to-speech technology for low-resource scenarios through Yannick Estève's TTS4ALL presentation, with each research talk followed by interactive discussion and Q&A sessions. Access detailed timestamps to navigate directly to specific presentations or discussions, making this an invaluable resource for researchers, students, and professionals interested in the latest developments in speech recognition, text-to-speech synthesis, language models, and audio processing technologies.