Patent Leaderboard
USPTO Patent Rankings Data through Dec 31, 2025
Home › Separating acoustic and linguistic information in neural transducer models for end-to-end speech recognition

Separating acoustic and linguistic information in neural transducer models for end-to-end speech recognition

US Patent 12394409 · Granted Aug 19, 2025

Assignee

Inventors

View full patent text on Google Patents →

More Patents by These Inventors

Domain adaptive speech recognition using artificial intelligence2025
Alternative soft label generation2025
External language model information integrated into neural transducer model2025
Accuracy of streaming RNN transducer2025
Learning unpaired multimodal feature matching for semi-supervised learning2025
Voice activity detection integration to improve automatic speech detection2025
Training of model for processing sequence data2024
Knowledge transfer between recurrent neural networks2024
Neural network for chemical compounds2024
Integrating text inputs for training and adapting neural network transducer ASR 2024
Customization of recurrent neural network transducers for speech recognition2024
Adding words to a prefix tree for improving speech recognition2024