An analog-AI chip for energy-efficient speech recognition and transcription.
Stefano AmbrogioPritish NarayananAtsuya OkazakiAndrea FasoliCharles MackinKohji HosokawaAkiyo NomuraTakeo YasudaAn ChenAlexander M. FrizMasatoshi IshiiJose LuquinYasuteru KohdaNicole SaulnierKevin BrewSamuel ChoiInjo OkTimothy PhilipVictor ChanMary Claire SilvestreIshtiaq AhsanVijay NarayananHsinyu TsaiGeoffrey W. BurrPublished in: Nat. (2023)
Keyphrases
- energy efficient
- speech recognition
- analog vlsi
- speech recognition technology
- speech recognition systems
- handwriting recognition
- wireless sensor networks
- artificial intelligence
- energy consumption
- sensor networks
- hidden markov models
- language model
- high speed
- base station
- pattern recognition
- speech signal
- automatic speech recognition
- speech processing
- speaker identification
- speech recognizer
- low cost
- energy efficiency
- speech synthesis
- machine learning
- students with learning disabilities
- noisy environments
- routing protocol
- intelligent systems
- speaker independent
- word error rate
- signal processing
- data mining
- power consumption
- wireless networks
- mobile phone