Model Architecture
Fig.1 Integrated vocal emotion conversion framework.
Speech Samples (Neutral to Anger)
Emotion Similarity Evaluation - German |
|
Source |
Converted |
Target (Reference) |
|
| |
|
|
|
|
Emotion Similarity Evaluation - English |
|
Source |
Converted |
Target (Reference) |
|
| |
|
|
|
|
Emotion Similarity Evaluation - Telugu |
|
Source |
Converted |
Target (Reference) |
|
| |
|
|
|
|
Speech Samples (Neutral to Fear)
Emotion Similarity Evaluation - German |
|
Source |
Converted |
Target (Reference) |
|
| |
|
|
|
|
Emotion Similarity Evaluation - English |
|
Source |
Converted |
Target (Reference) |
|
| |
|
|
|
|
Emotion Similarity Evaluation - Telugu |
|
Source |
Converted |
Target (Reference) |
|
| |
|
|
|
|
Speech Samples (Neutral to Happy)
Emotion Similarity Evaluation - German |
|
Source |
Converted |
Target (Reference) |
|
| |
|
|
|
|
Emotion Similarity Evaluation - English |
|
Source |
Converted |
Target (Reference) |
|
| |
|
|
|
|
Emotion Similarity Evaluation - Telugu |
|
Source |
Converted |
Target (Reference) |
|
| |
|
|
|
|