Update README.md
Browse files
README.md
CHANGED
|
@@ -27,15 +27,25 @@ dataset = load_dataset("PrathamOrgAI/ReadNet")
|
|
| 27 |
|
| 28 |
|
| 29 |
### Annotations
|
| 30 |
-
Annotations for the audio files are provided in the Transcribed Text Column
|
| 31 |
|
| 32 |
#### Annotation process
|
| 33 |
|
| 34 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 35 |
|
| 36 |
#### Who are the annotators?
|
| 37 |
|
| 38 |
-
|
| 39 |
|
| 40 |
### Personal and Sensitive Information
|
| 41 |
|
|
|
|
| 27 |
|
| 28 |
|
| 29 |
### Annotations
|
| 30 |
+
Annotations for the audio files are provided in the Transcribed Text Column. The annotation of audio files was done on an annotation portal which was developed inhouse by Pratham Education Foundation
|
| 31 |
|
| 32 |
#### Annotation process
|
| 33 |
|
| 34 |
+
A team of annotators were hired for the annotation of audio
|
| 35 |
+
samples in two languages. Training regarding the annotation
|
| 36 |
+
portal was provided to the annotators. A dry run for the an-
|
| 37 |
+
notation was done initially and the annotation was reviewed
|
| 38 |
+
and feedback was given to the annotators to remove incon-
|
| 39 |
+
sistencies and errors in annotation. Annotation guidelines are
|
| 40 |
+
provided in the annotation portal itself, in both languages for
|
| 41 |
+
the annotators reference. Since the scale of the data set is
|
| 42 |
+
pretty large, we have decided to annotate some portion of the
|
| 43 |
+
data twice and the remaining data once depending on the re-
|
| 44 |
+
sources
|
| 45 |
|
| 46 |
#### Who are the annotators?
|
| 47 |
|
| 48 |
+
The Annotation was done by experts from the ASER team
|
| 49 |
|
| 50 |
### Personal and Sensitive Information
|
| 51 |
|