MES-P: An Emotional Tonal Speech Dataset in Mandarin with Distal and Proximal Labels.
Zhongzhe XiaoYing ChenWeibei DouZhi TaoLiming ChenPublished in: IEEE Trans. Affect. Comput. (2022)
Keyphrases
- emotion recognition
- speech recognition
- broadcast news
- manually labeled
- weakly labeled
- emotional state
- audio visual
- training data
- spoken document retrieval
- speech synthesis
- facial expressions
- benchmark datasets
- sentiment analysis
- prosodic features
- training dataset
- images with ground truth
- automatic speech recognition
- text to speech
- human computer interaction
- speaker verification
- feature extraction
- test collection
- speech recognizer
- multi modal
- hidden markov models
- pattern recognition
- ground truth labels