Abstract
Fine-Tracker is a speech-based model of human speech recognition. While previous work has shown that Fine-Tracker is successful at modelling aspects of human spoken-word recognition, its speech recognition performance is not comparable to that of human performance, possibly due to suboptimal intermediate articulatory feature (AF) representations. This study investigates the effect of improved AF representations, obtained using a state-of-the-art deep convolutional network, on Fine-Tracker’s simulation and recognition performance: Although the improved AF quality resulted in improved speech recognition; it, surprisingly, did not lead to an improvement in Fine-Tracker’s simulation power.
| Original language | English |
|---|---|
| Title of host publication | Proceedings of the Machine Learning in Speech and Language Processing Workshop |
| Place of Publication | Hyderabad, India |
| Pages | 1-3 |
| Number of pages | 3 |
| Publication status | Published - 2018 |
| Event | Machine Learning in Speech and Language Processing Workshop - Google offices, Hyderabad, India Duration: 7 Sept 2018 → 7 Sept 2018 |
Workshop
| Workshop | Machine Learning in Speech and Language Processing Workshop |
|---|---|
| Country/Territory | India |
| City | Hyderabad |
| Period | 7/09/18 → 7/09/18 |
Keywords
- Convolutional Neural Network
- spoken-word recognition
- computational modelling
- articulatory features