Speech Recognition (epo) Patents (Class 704/E15.001)
- Using distance or distortion measures between unknown speech and reference templates (EPO) (Class 704/E15.015)
- Using dynamic programming techniques, e.g., Dynamic Time Warping (DTW), etc. (EPO) (Class 704/E15.016)
- Using artificial neural networks (EPO) (Class 704/E15.017)
- Using natural language modeling (EPO) (Class 704/E15.018)
- Using statistical models, e.g., Hidden Markov Models (HMMs), etc. (EPO) (Class 704/E15.027)
- Recognition networks (EPO) (Class 704/E15.038)
- Distributed recognition, e.g., in client-server systems for mobile phones or network applications, etc. (EPO) (Class 704/E15.047)
- Memory allocation or algorithm optimization to reduce hardware requirements (EPO) (Class 704/E15.048)
- Multiple recognizers used in sequence or in parallel; corresponding voting or score combination systems (EPO) (Class 704/E15.049)
- Recognizers for parallel processing (EPO) (Class 704/E15.05)