VoiceFilter-Lite: Streaming Targeted Voice Separation for On-Device Speech Recognition.
Quan WangIgnacio Lopez-MorenoMert SaglamKevin W. WilsonAlan ChiaoRenjie LiuYanzhang HeWei LiJason PelecanosMarily NikaAlexander GruensteinPublished in: CoRR (2020)
Keyphrases
- speech recognition
- speech recognition errors
- speech synthesis
- voice activity detection
- hidden markov models
- pattern recognition
- speech processing
- automatic speech recognition
- language model
- speech recognition systems
- speech recognizer
- speech understanding
- speech signal
- keyword spotting
- noisy environments
- text to speech
- speech recognition technology
- speech recognizers
- mel frequency cepstral coefficients
- speaker independent
- cepstral coefficients
- speaker identification
- video data
- vocal tract
- speaker dependent
- speech retrieval
- isolated word