🚨Attention audio researchers!
We have released a new dataset, CASTELLA, consisting of long audio recordings, captions, and temporal boundaries.
🔍 Paper: https://t.co/sDV0ZlYG9v
🌐 Project page: https://t.co/4KJJvAHMas
Today, I’ll be presenting “BitTTS: Highly Compact Text-to-Speech Using 1.58-bit Quantization and Weight Indexing” at Port 1C (15:10-15:30)
https://t.co/h10uY5mUbA
Today, I will have a presentation titled “DnR-nonverbal: Cinematic Audio Source Separation Dataset Containing Non-Verbal Sounds” at Port 1A, (09:50-10:10). #Interspeech2025
My co-worker, Kawamura-kun will also have a presentation titled “BitTTS: Highly Compact Text-to-Speech Using 1.58-bit Quantization and Weight Indexing” at Port 1C (15:10-15:30). #Interspeech2025
Today, I will have a presentation titled “DnR-nonverbal: Cinematic Audio Source Separation Dataset Containing Non-Verbal Sounds” at Port 1A, (09:50-10:10). #Interspeech2025