Evaluating the Reliability of Acoustic Speech Embeddings.
Robin AlgayresMohamed Salah ZaïemBenoît SagotEmmanuel DupouxPublished in: INTERSPEECH (2020)
Keyphrases
- speech recognition
- speech sounds
- speech recognition systems
- acoustic features
- automatic speech recognition
- emotional speech
- speech signal
- acoustic signal
- acoustic models
- source localization
- speaker recognition
- formant frequencies
- endpoint detection
- speech recognizers
- noisy speech
- prosodic features
- speaker verification
- highly reliable
- broadcast news
- audio visual
- dimensionality reduction
- text to speech
- pattern recognition
- speech recognizer
- music information retrieval
- emotion recognition
- euclidean space
- vector space
- multi modal
- low dimensional