Improved Training Strategies for End-to-End Speech Recognition in Digital Voice Assistants.
Hitesh TulsianiAshtosh SapruHarish ArsikereSurabhi PunjabiSri GarimellaPublished in: INTERSPEECH (2020)
Keyphrases
- end to end
- speech recognition
- wall street journal corpus
- speech synthesis
- isolated word
- speech recognition errors
- hidden markov models
- language model
- acoustic models
- speech recognizer
- speech recognition systems
- automatic speech recognition
- voice activity detection
- digital video library
- speech processing
- noisy environments
- speech signal
- congestion control
- speech recognition technology
- speaker identification
- pattern recognition
- training process
- speaker dependent
- speech recognizers
- speaker diarization