Datasets:
Update README.md
Browse files
README.md
CHANGED
|
@@ -3,4 +3,19 @@ license: gpl-3.0
|
|
| 3 |
language:
|
| 4 |
- en
|
| 5 |
pretty_name: MELD Preprocessed
|
| 6 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
language:
|
| 4 |
- en
|
| 5 |
pretty_name: MELD Preprocessed
|
| 6 |
+
---
|
| 7 |
+
|
| 8 |
+
# MELD Preprocessed for SER
|
| 9 |
+
|
| 10 |
+
This dataset is the manually preprocessed audio only version of [MELD](https://affective-meld.github.io/), only audio IDs, utterance transcriptions, dialogue IDs and Utterance IDs were extracted.
|
| 11 |
+
|
| 12 |
+
```
|
| 13 |
+
S. Poria, D. Hazarika, N. Majumder, G. Naik, R. Mihalcea,
|
| 14 |
+
E. Cambria. MELD: A Multimodal Multi-Party Dataset
|
| 15 |
+
for Emotion Recognition in Conversation. (2018)
|
| 16 |
+
```
|
| 17 |
+
```
|
| 18 |
+
Chen, S.Y., Hsu, C.C., Kuo, C.C. and Ku, L.W.
|
| 19 |
+
EmotionLines: An Emotion Corpus of Multi-Party
|
| 20 |
+
Conversations. arXiv preprint arXiv:1802.08379 (2018).
|
| 21 |
+
```
|