Audio2Face: Generating Speech/Face Animation from Single Audio with Attention-Based Bidirectional LSTM Networks.
Guanzhong TianYi YuanYong LiuPublished in: ICME Workshops (2019)
Keyphrases
- audio visual
- multimedia
- human faces
- recognition engine
- audio stream
- face images
- audio signals
- emotion recognition
- broadcast news
- speaker identification
- multimodal fusion
- computer graphics
- facial images
- automatic transcription
- social networks
- cepstral features
- visual speech
- audio video
- audio features
- facial expressions
- text to speech
- speech synthesis
- multimedia information
- speech recognition
- digital audio
- signal processing
- spoken documents
- hidden markov models
- image sequences