Sumformer: A Linear-Complexity Alternative to Self-Attention for Speech Recognition.
Titouan ParcolletRogier van DalenShucong ZhangSourav BhattacharyaPublished in: CoRR (2023)
Keyphrases
- speech recognition
- linear complexity
- hidden markov models
- language model
- speech processing
- speech synthesis
- speech signal
- speech recognizer
- keyword spotting
- speaker identification
- speech recognition technology
- automatic speech recognition
- noisy environments
- pattern recognition
- speech understanding
- speech recognition systems
- speaker independent
- speech recognizers
- handwriting recognition
- speaker diarization
- n gram
- mobile devices
- speaker dependent
- computer vision
- neural network