VoiceFilter: Targeted Voice Separation by Speaker-Conditioned Spectrogram Masking.
Quan WangHannah MuckenhirnKevin W. WilsonPrashant SridharZelin WuJohn R. HersheyRif A. SaurousRon J. WeissYe JiaIgnacio Lopez-MorenoPublished in: CoRR (2018)
Keyphrases
- automatic speech recognition
- speech signal
- synthesized speech
- mel frequency cepstral coefficients
- speech sounds
- speaker recognition
- speech recognition
- pattern analysis
- prosodic features
- speaker verification
- human visual system
- speaker identification
- fundamental frequency
- emotion recognition
- audio visual
- voice and data services
- wigner distribution
- hidden markov models
- speaker diarization
- text to speech
- image analysis
- speech synthesis
- acoustic features
- speaker dependent
- broadcast news
- sound source
- voice activity detection