Datasets:
Tasks:
Automatic Speech Recognition
Languages:
Maltese
Multilinguality:
monolingual
Size Categories:
n<1K
Language Creators:
other
Annotations Creators:
expert-generated
Source Datasets:
original
License:
Commit
•
6971cf9
1
Parent(s):
0ffb90d
Update README.md
Browse files
README.md
CHANGED
|
@@ -61,12 +61,12 @@ The MASRI-DEV CORPUS was created out of YouTube videos belonging to the channel
|
|
| 61 |
The MASRI-DEV contains only the validation split:
|
| 62 |
```python
|
| 63 |
from datasets import load_dataset
|
| 64 |
-
masri_dev = load_dataset("
|
| 65 |
```
|
| 66 |
It is also valid to do:
|
| 67 |
```python
|
| 68 |
from datasets import load_dataset
|
| 69 |
-
masri_dev = load_dataset("
|
| 70 |
```
|
| 71 |
### Supported Tasks
|
| 72 |
automatic-speech-recognition: The dataset can be used as a development/validation corpus when training a model for Automatic Speech Recognition (ASR). The model is presented with an audio file and asked to transcribe the audio file to written text. The most common evaluation metric is the word error rate (WER).
|
|
|
|
| 61 |
The MASRI-DEV contains only the validation split:
|
| 62 |
```python
|
| 63 |
from datasets import load_dataset
|
| 64 |
+
masri_dev = load_dataset("MLRS/masri_dev")
|
| 65 |
```
|
| 66 |
It is also valid to do:
|
| 67 |
```python
|
| 68 |
from datasets import load_dataset
|
| 69 |
+
masri_dev = load_dataset("MLRS/masri_dev",split="validation")
|
| 70 |
```
|
| 71 |
### Supported Tasks
|
| 72 |
automatic-speech-recognition: The dataset can be used as a development/validation corpus when training a model for Automatic Speech Recognition (ASR). The model is presented with an audio file and asked to transcribe the audio file to written text. The most common evaluation metric is the word error rate (WER).
|