A Hybrid Approach to Acoustic Scene Classification Based on Universal Acoustic Models.
Xue BaiJun DuZi-Rui WangChin-Hui LeePublished in: INTERSPEECH (2019)
Keyphrases
- scene classification
- acoustic models
- speech recognition
- hidden markov models
- speech recognizer
- broadcast news
- object recognition
- natural scenes
- automatic speech recognition
- biologically inspired
- image classification
- speaker independent
- discriminative training
- visual words
- image representation
- speech recognition systems
- bag of features
- speech signal
- spoken language
- pattern recognition
- multiscale
- training data
- speaker identification
- image segmentation