A 50μW 4-channel 83dBA-SNDR Speech Recognition Front-End with Adaptive Beamforming and Feature Extraction.
Taewook KangSeungjong LeeMohammad HaghigatDarren AbramsonMichael P. FlynnPublished in: CICC (2021)
Keyphrases
- speech recognition
- feature extraction
- pattern recognition
- mimo ofdm systems
- speaker identification
- hidden markov models
- cepstral coefficients
- speech signal
- speech recognizer
- speech processing
- frequency domain
- noisy environments
- speech synthesis
- language model
- automatic speech recognition
- speech recognition technology
- speech enhancement
- feature selection
- speech recognition systems
- speech understanding
- image processing
- principal component analysis
- extracted features
- signal processing
- speech recognition errors
- isolated word
- feature space
- computer vision
- feature vectors
- speaker dependent
- wavelet transform