Patent Leaderboard
USPTO Patent Rankings Data through Dec 31, 2025
Home › Enhanced automatic speech recognition using mapping between unsupervised and supervised speech model parameters trained on same acoustic training data

Enhanced automatic speech recognition using mapping between unsupervised and supervised speech model parameters trained on same acoustic training data

US Patent 8306819 · Granted Nov 6, 2012

Estimated economic value: $20,259,000 ⓘ

Assignee

Inventors

View full patent text on Google Patents →

More Patents by These Inventors

Core group memory processing with multi-precision weight packing2025
Chiplet based artificial intelligence accelerators and configuration methods2025
Memory processing unit core architectures2025
Generating and using text-to-speech data for speech recognition models2025
Communication method, related computing system and storage medium for selecting 2024
Machine learning model with depth processing units2024
Dynamic combination of acoustic model states2024
Speaker adaptation for attention-based encoder-decoder2024
Augmented training data for end-to-end models2024
Condition-invariant feature extraction network2023
On-device custom wake word detection2023
Wake word selection assistance architectures and methods2023