Leveraging Sequence-to-Sequence Speech Synthesis for Enhancing Acoustic-to-Word Speech Recognition.
Masato MimuraSei UenoHirofumi InagumaShinsuke SakaiTatsuya KawaharaPublished in: SLT (2018)
Keyphrases
- speech recognition
- speech synthesis
- prosodic features
- speech recognition systems
- speech recognizers
- speech recognizer
- text to speech
- vocal tract
- hidden markov models
- pattern recognition
- speech signal
- keyword spotting
- speaker independent
- language model
- wall street journal corpus
- automatic speech recognition
- speech processing
- speaker identification
- speech recognition technology
- noisy speech
- neural network
- acoustic models
- n gram
- maximum likelihood
- machine learning