Patent Leaderboard
USPTO Patent Rankings Data through Dec 31, 2025
Home › Separating speech by source in audio recordings by predicting isolated audio signals conditioned on speaker representations

Separating speech by source in audio recordings by predicting isolated audio signals conditioned on speaker representations

US Patent 12236970 · Granted Feb 25, 2025

Assignee

Inventors

View full patent text on Google Patents →

More Patents by These Inventors

Learned audio frontend machine learning model for audio understanding2025
Generating audio using auto-regressive generative neural networks2025
Generating coded data representations using neural networks and vector quantizer2025
Spatial audio recording from home assistant devices2025
Generating audio using auto-regressive generative neural networks2024
Generating audio using auto-regressive generative neural networks2024
Compressing audio waveforms using neural networks and vector quantizers2024
Generating audio using auto-regressive generative neural networks2024
End-to-end speech diarization via iterative speaker embedding2024
End-to-end speech diarization via iterative speaker embedding2024
Compressing audio waveforms using neural networks and vector quantizers2023
Separating speech by source in audio recordings by predicting isolated audio sig2022