Enhanced stability prediction for incrementally generated speech recognition hypotheses based on an age of a hypothesis
Patent №
US 8,909,512
Granted
2014-12-09
Filed 2012
Owner
GOOGLE INC.
AI components
3
nlp · speech · hardware
Assignment
Recorded
Dataset
AIPD
2023_r1 edition
Application
13461033
Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for predicting the stability of speech recognition results. In one aspect, a method includes determining a length of time, or a number of occasions, in which a word has remained in an incremental speech recognizer's top hypothesis, and assigning a stability metric to the word based on the length of time or number of occasions.
Natural languageSpeechAI hardwareG03F 7/70708G10L 15/08G03F 7/707G03F 7/70875G03F 7/70908H10P 72/72G10L 2015/223
AI classification
Natural language1.00
Speech1.00
AI hardware0.86
Vision0.21
Planning0.16
Knowledge representation0.11
Machine learning0.04
Evolutionary computation0.00
Ownership
GOOGLE INC.
assignment · 291470159
Assignors
MCGRAW, IAN C., GRUENSTEIN, ALEXANDER H.
On an employer assignment, the assignors are typically the inventors.
From the same owner
ADAPTIVE RANKING OF NAVIGATION SUGGESTIONSUS 12,694,045 · 2026Multi-Image SearchUS 12,694,053 · 2026GENERATING MULTI-MODAL RESPONSE(S) THROUGH UTILIZATION OF LARGE LANGUAGE MODEL(S)US 12,694,231 · 2026MULTI-STAGE COMPUTATIONALLY EFFICIENT NEURAL NETWORK INFERENCEUS 12,694,268 · 2026HIGH-RESOLUTION MULTIVIEW-CONSISTENT RENDERING AND ALPHA MATTING FROM SPARSE VIEWSUS 12,694,480 · 2026Nonlinear Peri-Codec Optimization For Image And Video CodingUS 12,694,572 · 2026OBJECT OUTLINE GENERATION FROM OVERHEAD IMAGERY USING ACTION SEQUENCE PREDICTIONUS 12,694,670 · 2026CONTEXT-BASED SPEAKER COUNTER FOR A SPEAKER DIARIZATION SYSTEMUS 12,694,708 · 2026