tobiolatunji
commited on
Commit
·
50745c8
1
Parent(s):
0b720c8
update readme with config for individual accents
Browse files
README.md
CHANGED
@@ -101,9 +101,9 @@ afrispeech = load_dataset("tobiolatunji/afrispeech-200", "all")
|
|
101 |
|
102 |
The entire dataset is ~120GB and may take about 2hrs to download depending on internet speed/bandwidth. If you have disk space or bandwidth limitations, you can use `streaming` mode described below to work with smaller subsets of the data.
|
103 |
|
104 |
-
|
105 |
|
106 |
-
For example, to download the isizulu config, simply specify the corresponding
|
107 |
```python
|
108 |
from datasets import load_dataset
|
109 |
|
@@ -141,10 +141,13 @@ afrispeech = load_dataset("tobiolatunji/afrispeech-200", "isizulu", split="train
|
|
141 |
dataloader = DataLoader(afrispeech, batch_size=32)
|
142 |
```
|
143 |
|
|
|
|
|
|
|
144 |
|
145 |
### Fine-tuning Colab tutorial
|
146 |
|
147 |
-
To walk through a complete colab tutorial that finetunes a wav2vec2 model on the afrispeech-200 dataset with `transformers`, take a look at this colab notebook [afrispeech/wav2vec2-colab-tutorial](https://colab.research.google.com/drive/1uZYew6pcgN6UE6sFDLohxD_HKivvDXzD?usp=sharing
|
148 |
|
149 |
|
150 |
### Supported Tasks and Leaderboards
|
|
|
101 |
|
102 |
The entire dataset is ~120GB and may take about 2hrs to download depending on internet speed/bandwidth. If you have disk space or bandwidth limitations, you can use `streaming` mode described below to work with smaller subsets of the data.
|
103 |
|
104 |
+
Alterntively you are able to pass a config to the `load_dataset` function and download only a subset of the data corresponding to a specific accent of interest. The example provided below is `isizulu`.
|
105 |
|
106 |
+
For example, to download the isizulu config, simply specify the corresponding accent config name. The list of supported accents is provided in the `accent list` section below:
|
107 |
```python
|
108 |
from datasets import load_dataset
|
109 |
|
|
|
141 |
dataloader = DataLoader(afrispeech, batch_size=32)
|
142 |
```
|
143 |
|
144 |
+
### Caveats
|
145 |
+
Note that till the end of the ongoing [AfriSpeech ASR Challenge event](https://zindi.africa/competitions/intron-afrispeech-200-automatic-speech-recognition-challenge) (Feb - May 2023), the transcripts in the validation set are hidden and the test set will be unreleased till May 19, 2023.
|
146 |
+
|
147 |
|
148 |
### Fine-tuning Colab tutorial
|
149 |
|
150 |
+
To walk through a complete colab tutorial that finetunes a wav2vec2 model on the afrispeech-200 dataset with `transformers`, take a look at this colab notebook [afrispeech/wav2vec2-colab-tutorial](https://colab.research.google.com/drive/1uZYew6pcgN6UE6sFDLohxD_HKivvDXzD?usp=sharing).
|
151 |
|
152 |
|
153 |
### Supported Tasks and Leaderboards
|