A 16-nm SoC for Noise-Robust Speech and NLP Edge AI Inference With Bayesian Sound Source Separation and Attention-Based DNNs.
Thierry TambeEn-Yu YangGlenn G. KoYuji ChaiColeman HooperMarco DonatoPaul N. WhatmoughAlexander M. RushDavid BrooksGu-Yeon WeiPublished in: IEEE J. Solid State Circuits (2023)
Keyphrases
- source separation
- noisy environments
- speech recognition
- bayesian networks
- audio features
- bayesian inference
- natural language processing
- blind source separation
- edge detection
- missing data
- independent component analysis
- single channel
- information extraction
- machine learning
- denoising
- text mining
- signal to noise ratio
- noise reduction
- visual features
- focus of attention