Update README.md
Browse files
README.md
CHANGED
|
@@ -26,15 +26,22 @@ It achieves the following results on the evaluation set:
|
|
| 26 |
|
| 27 |
## Model description
|
| 28 |
|
| 29 |
-
|
| 30 |
|
| 31 |
## Intended uses & limitations
|
| 32 |
|
| 33 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 34 |
|
| 35 |
## Training and evaluation data
|
| 36 |
|
| 37 |
-
|
|
|
|
|
|
|
|
|
|
| 38 |
|
| 39 |
## Training procedure
|
| 40 |
|
|
|
|
| 26 |
|
| 27 |
## Model description
|
| 28 |
|
| 29 |
+
This model is designed to convert Arabic text to Arabic gloss, where the gloss is an intermediary representation used in sign language translation.
|
| 30 |
|
| 31 |
## Intended uses & limitations
|
| 32 |
|
| 33 |
+
-Source: Arabic Text
|
| 34 |
+
-Target: Arabic Gloss
|
| 35 |
+
-Domain: Islamic Friday Sermon trasulation from spoken text to ArSL gloss
|
| 36 |
+
|
| 37 |
+
Intended Uses: This model can be used for educational purposes, research in Arabic sign language translation (ArSL), and as a tool for developing applications for the deaf community in the Islamic Friday Sermon translation domain. Limitations: The model's performance is highly dependent on the quality and domain of the input data. It may not perform well on out-of-domain text.
|
| 38 |
|
| 39 |
## Training and evaluation data
|
| 40 |
|
| 41 |
+
The dataset was split into training and test sets as follows:
|
| 42 |
+
-Training set: 6052 examples ... 70%
|
| 43 |
+
-Evaluation set: 1297 examples ... 15%
|
| 44 |
+
-Test set: 1297 examples ... 15%
|
| 45 |
|
| 46 |
## Training procedure
|
| 47 |
|