---
license: mit
task_categories:
- audio-to-audio
language:
- es
- en
tags:
- Football (soccer)
- audio
- Football
size_categories: n<1K
---
# my-fancy-dataset
---

## Description

Speech dataset from English and Spanish football commentary. Data was sourced from BBC News and ITV News football commentary during the World Cup final. The dataset was used to train an LLM called 'Valyu-LLM-Football'. The dataset contains 100,000,000 audio samples. The annotation process involved manual human annotation.
### Dataset Size

**Audio**

- Number of Samples: 100000000

## Characteristics

| Field | Value |
|-------|-------|
| Topics | Football (soccer), Football |
| Languages | Spanish, English |
| Task Categories | audio-to-audio |
| Modalities | audio |





![Characteristics Image](https://datacard-provenance-images.s3.eu-west-2.amazonaws.com/963Eum3g5gRI41gt1r5YL0saqyQ2/my-fancy-dataset/characteristics-screenshot.png)


## Provenance

| Field               | Details                      |
|---------------------|------------------------------|
| Creators            | yamamuchiiii                   |
| Derived from the following sources| BBC News, ITV News       |
| Synthetic data  | False         |

![Provenance Image](https://datacard-provenance-images.s3.eu-west-2.amazonaws.com/963Eum3g5gRI41gt1r5YL0saqyQ2/my-fancy-dataset/provenance-screenshot.png)


- This dataset was used to train Valyu-LLM-Football
## Licenses
The dataset is licensed under:
- MIT License (OPEN)
## Annotation Processes
manual human annotation



Documentation generated by ![Characteristics Image](https://datacard-provenance-images.s3.eu-west-2.amazonaws.com/generated-by.png)