CHiVE: Examples for randomly sampled embeddings
Figure 5: Audio for randomly generated log F0 curves
These examples show a random sample of the varied prosodies for the same sentence.
| 1 | |
| 2 | |
| 3 | |
| 4 | |
| 5 | |
| 6 | |
| 7 | |
| 8 |
| 1 | |
| 2 | |
| 3 | |
| 4 | |
| 5 | |
| 6 | |
| 7 | |
| 8 |
This page contains audio samples that go with Modelling Intonation in Spectrograms for Neural Vocoder based Text-to-Speech, Wan, V., Shen, J., Silen, H. and Clark, R. A., Proceedings of Speech Prosody 2020.