AnnoTheia: A Semi-Automatic Annotation Toolkit for Audio-Visual Speech Technologies.
José-M. Acosta-TrianaDavid Gimeno-GómezCarlos D. Martínez-HinarejosPublished in: CoRR (2024)
Keyphrases
- automatic annotation
- visual speech
- visual information
- hidden markov models
- content based retrieval
- speaker identification
- audio signals
- image annotation
- low level features
- semantic annotation
- noisy environments
- acoustic features
- broadcast news
- video signals
- gaussian mixture model
- audio signal
- visual concepts
- multimedia
- similarity measure
- visual features
- audio visual
- object recognition
- computer vision
- video sequences
- speech signal
- image sequences
- training set
- eye movements