vonewman commited on
Commit
4fed4dd
·
verified ·
1 Parent(s): 4a76430

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +49 -28
README.md CHANGED
@@ -1,55 +1,76 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  # Urban Bus Wolof Speech Dataset
2
 
3
- Ce dataset contient des enregistrements audio et leurs transcriptions en wolof, liés au domaine du transport urbain par bus. L'objectif est de faciliter le développement de modèles de Reconnaissance Automatique de la Parole (ASR) pour aider les personnes illettrées à utiliser des applications existantes pour trouver quel bus prendre pour atteindre leur destination, sans avoir besoin de savoir lire ou écrire.
4
 
5
- ## Description du Dataset
6
 
7
- - **Langue** : Wolof
8
- - **Domaine** : Transport urbain
9
- - **Type de données** : Audio et transcription
10
- - **Format des fichiers audio** : MP3
11
- - **Taux d'échantillonnage** : 16 kHz
12
- - **Nombre total d'exemples** : 6 683
13
- - **Entraînement** : 5 346 exemples (80%)
14
- - **Test** : 1 337 exemples (20%)
15
 
16
- ## Structure du Dataset
17
 
18
- ### Caractéristiques (Features)
19
 
20
- - `audio` : Fichier audio contenant l'enregistrement de la parole en wolof.
21
- - **Format** : MP3
22
- - **Taux d'échantillonnage** : 16 kHz
23
- - `sentence` : Transcription textuelle de l'audio en wolof.
24
 
25
  ### Splits
26
 
27
- Le dataset est divisé en deux splits :
28
 
29
- | Split | Nombre d'exemples |
30
- |--------|-------------------|
31
- | Train | 5 346 |
32
- | Test | 1 337 |
33
 
34
- ## Exemple d'utilisation
35
 
36
- Voici comment charger et utiliser ce dataset avec la bibliothèque 🤗 Datasets :
37
 
38
  ```python
39
  from datasets import load_dataset
40
 
41
- # Charger le dataset
42
  dataset = load_dataset("vonewman/urban-bus-wolof")
43
 
44
- # Accéder à un exemple du split 'train'
45
  print(dataset['train'][0])
46
 
47
- # Sortie attendue :
48
  # {
49
  # 'audio': {
50
- # 'path': '.../train/audio/<nom_du_fichier>.mp3',
51
  # 'array': array([...]),
52
  # 'sampling_rate': 16000
53
  # },
54
- # 'sentence': 'Transcription associée à l\'audio en wolof'
55
  # }
 
1
+ ---
2
+ annotations_creators:
3
+ - expert-generated
4
+ language_creators:
5
+ - found
6
+ language:
7
+ - wo
8
+ license:
9
+ - cc-by-4.0
10
+ multilinguality:
11
+ - monolingual
12
+ size_categories:
13
+ - 1K<n<10K
14
+ source_datasets:
15
+ - original
16
+ task_categories:
17
+ - automatic-speech-recognition
18
+ task_ids:
19
+ - speech-recognition
20
+ ---
21
+
22
  # Urban Bus Wolof Speech Dataset
23
 
24
+ This dataset contains audio recordings and their transcriptions in Wolof, related to urban bus transportation. The goal is to facilitate the development of Automatic Speech Recognition (ASR) models to help illiterate people use existing apps to find which bus they can take to reach their destination without needing to know how to read or write.
25
 
26
+ ## Dataset Description
27
 
28
+ - **Language**: Wolof (`wo`)
29
+ - **Domain**: Urban transportation
30
+ - **Data Type**: Audio recordings and transcriptions
31
+ - **Audio File Format**: MP3
32
+ - **Sampling Rate**: 16 kHz
33
+ - **Total Examples**: 6,683
34
+ - **Training Set**: 5,346 examples (80%)
35
+ - **Test Set**: 1,337 examples (20%)
36
 
37
+ ## Dataset Structure
38
 
39
+ ### Features
40
 
41
+ - `audio`: An audio file containing the speech in Wolof.
42
+ - **Format**: MP3
43
+ - **Sampling Rate**: 16 kHz
44
+ - `sentence`: The textual transcription of the audio in Wolof.
45
 
46
  ### Splits
47
 
48
+ The dataset is divided into two splits:
49
 
50
+ | Split | Number of Examples |
51
+ |--------|---------------------|
52
+ | Train | 5,346 |
53
+ | Test | 1,337 |
54
 
55
+ ## Example Usage
56
 
57
+ Here's how to load and use this dataset with the 🤗 Datasets library:
58
 
59
  ```python
60
  from datasets import load_dataset
61
 
62
+ # Load the dataset
63
  dataset = load_dataset("vonewman/urban-bus-wolof")
64
 
65
+ # Access an example from the 'train' split
66
  print(dataset['train'][0])
67
 
68
+ # Expected output:
69
  # {
70
  # 'audio': {
71
+ # 'path': '.../train/audio/<audio_file>.mp3',
72
  # 'array': array([...]),
73
  # 'sampling_rate': 16000
74
  # },
75
+ # 'sentence': 'Transcription of the audio in Wolof'
76
  # }