Toward Better Speaker Embeddings: Automated Collection of Speech Samples From Unknown Distinct Speakers.
Minh PhamZeqian LiJacob WhitehillPublished in: ICASSP (2020)
Keyphrases
- speech recognition
- speaker dependent
- automatic speech recognition
- speaker identification
- speech signal
- speaker adaptation
- speech synthesis
- speaker recognition
- phoneme recognition
- speaker verification
- audio visual
- speech recognizer
- hidden markov models
- data sets
- pattern recognition
- speaker independent
- noisy environments
- training set
- semi automated
- vocal tract
- speaker diarization
- low dimensional
- recognition engine
- speech recognition systems
- language model
- automatic speech recognition systems
- automated analysis
- random sample
- training samples
- vector space
- maximum likelihood
- acoustic features
- gaussian mixture model
- high dimensional data
- prosodic features
- dimensionality reduction
- speech sounds
- information retrieval