---
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
  - split: test
    path: data/test-*
dataset_info:
  features:
  - name: audio
    dtype: audio
  - name: client_id
    dtype: string
  - name: transcription
    dtype: string
  - name: up_votes
    dtype: int64
  - name: down_votes
    dtype: int64
  - name: age
    dtype: string
  - name: gender
    dtype: string
  - name: accents
    dtype: string
  - name: variant
    dtype: float64
  - name: locale
    dtype: string
  - name: segment
    dtype: float64
  splits:
  - name: train
    num_bytes: 133629462.356
    num_examples: 5324
  - name: validation
    num_bytes: 1039373547.526
    num_examples: 42418
  - name: test
    num_bytes: 107435663.014
    num_examples: 4031
  download_size: 1266451644
  dataset_size: 1280438672.896
task_categories:
- conversational
- translation
language:
- ur
tags:
- code
---

# Dataset Card for AudioDataset-15

## Table of Contents
- [Table of Contents](#table-of-contents)
- [Dataset Description](#dataset-description)
  - [Dataset Summary](#dataset-summary)
  - [Languages](#languages)
- [Dataset Structure](#dataset-structure)
  - [Data Instances](#data-instances)
  - [Data Fields](#data-fields)
  - [Data Splits](#data-splits)

## Dataset Description

### Dataset Summary

The dataset in question is an audio dataset consisting of recordings in the Urdu language. It has been sourced from Mozilla's Common Voice, a publicly available voice dataset that relies on the contributions of volunteers from various parts of the world. The primary purpose of this dataset is to support the development of voice applications by providing a valuable resource for training machine learning models.

The dataset's intended use is to facilitate voice-to-text conversion in the Urdu language. By utilizing this dataset, researchers, developers, and anyone interested in voice technology can train models that accurately convert spoken Urdu words into written text. This can have significant applications in various domains, such as speech recognition, transcription services, language learning tools, and more.

### Languages

The dataset consists of audio recordings in the Urdu language. Urdu is a language primarily spoken in Pakistan and parts of India. It is one of the 22 officially recognized languages in India and is also widely spoken by the Pakistani diaspora around the world.

The dataset is primarily focused on spoken Urdu, which encompasses a wide range of topics and genres. It is important to note that the dataset's content may vary, covering conversations, speeches, interviews, narratives, and other forms of vocal communication in the Urdu language.

## Dataset Structure

### Data Instances

{
  "client_id": "0c9690e5a2d1bb3ce418954a2b70acae53153708f6c3a21c9e8fe7e3912d97ba805ace5091772c8d4e16dc07fc906ca4956335b87821c244eee8129a15fcb0cf",
  "file_name": "data/test/common_voice_ur_26641307.mp3",
  "transcription": "تو ان کے حلاج مدلوں کا کیا حال ہے؟",
  "up_votes": 2,
  "down_votes": 0,
  "age": "twenties",
  "gender": "female",
  "accent": "",
  "locale": "ur",
  "segment": ""
}

### Data Fields
<li>client_id: A unique identifier for the client or contributor who provided the recording. (Data Type: String)</li>
<li>file_name: The file name or path of the audio file. (Data Type: String)</li>
<li>transcription: The transcription of the spoken content in the Urdu language. (Data Type: String)</li>
<li>up_votes: The number of upvotes received for the recording. (Data Type: Integer)</li>
<li>down_votes: The number of downvotes received for the recording. (Data Type: Integer)</li>
<li>age: The age group of the speaker. (Data Type: String)</li>
<li>gender: The gender of the speaker. (Data Type: String)</li>
<li>accent: The accent of the speaker, if applicable. (Data Type: String)</li>
<li>locale: The locale or language code, which is "ur" for Urdu in this case. (Data Type: String)</li>
<li>segment: Additional segment information, if available. (Data Type: String)</li>

### Data Splits
The dataset is divided into three splits: train, test, and validation. The training set is used to train the model, the validation set is used to tune hyperparameters and evaluate model performance during training, and the test set is used to evaluate the final model's performance after training.
|                         | train | validation | test  |
|-------------------------|------:|-----------:|-----: |
| Amount                  |  5324  |     42418     |  4031  |