Blind Extraction of Moving Audio Source in a Challenging Environment Supported by Speaker Identification Via X-Vectors.
Jirí MálekJakub JanskýTomás KounovskýZbynek KoldovskýJindrich ZdánskýPublished in: ICASSP (2021)
Keyphrases
- speaker identification
- noisy environments
- gaussian mixture model
- speech processing
- speech recognition
- speech signal
- feature extraction
- speaker recognition
- audio signals
- broadcast news
- audio signal
- information extraction
- feature vectors
- high dimensional
- multi modal
- automatic speech recognition
- audio features
- speaker diarization
- computer vision
- single point
- noise reduction
- visual information
- low level
- multimedia