More is Less: Domain-Specific Speech Recognition Microprocessor Using One-Dimensional Convolutional Recurrent Neural Network.
Bo LiuHao CaiZilong ZhangXiaoling DingZiyu WangYu GongWeiqiang LiuJinjiang YangZhen WangJun YangPublished in: IEEE Trans. Circuits Syst. I Regul. Pap. (2022)
Keyphrases
- speech recognition
- recurrent neural networks
- domain specific
- recurrent networks
- hidden markov models
- feed forward
- neural network
- pattern recognition
- language model
- echo state networks
- speech recognizer
- complex valued
- speech processing
- reservoir computing
- automatic speech recognition
- speech signal
- artificial neural networks
- speech understanding
- deep learning
- speech synthesis
- hidden layer
- noisy environments
- speech recognition technology
- speech recognition systems
- speaker dependent
- artificial intelligence
- speech retrieval
- speaker independent
- neural network structure
- keyword spotting
- isolated word
- non stationary