README.md

---
license: cc-by-4.0
task_categories:
- audio-classification
language:
- de
- en
- es
- fr
- it
- nl
- pl
- sv
tags:
- speech
- speech-classifiation
- text-to-speech
- spoofing
- multilingualism

pretty_name: FLEURS-HS
size_categories:
- 10K<n<100K
---

# FLEURS-HS

An extension of the [FLEURS](https://huggingface.co/datasets/google/fleurs) dataset for synthetic speech detection using text-to-speech, featured in the paper **Synthetic speech detection with Wav2Vec 2.0 in various language settings**.

This dataset is 1 of 3 used in the paper, the others being:
- [FLEURS-HS VITS](https://huggingface.co/datasets/realnetworks-kontxt/fleurs-hs-vits)
  - test set containing (generally) more difficult synthetic samples
  - separated due to different licensing
- [ARCTIC-HS](https://huggingface.co/datasets/realnetworks-kontxt/arctic-hs)
  - extension of the [CMU_ARCTIC](http://festvox.org/cmu_arctic/) and [L2-ARCTIC](https://psi.engr.tamu.edu/l2-arctic-corpus/) sets in a similar manner

## Dataset Details

### Dataset Description

The dataset features 8 languages originally seen in FLEURS:

- German
- English
- Spanish
- French
- Italian
- Dutch
- Polish
- Swedish

The original FLEURS samples are used as `human` samples, while `synthetic` samples are generated using:

- [Google Cloud Text-To-Speech](https://cloud.google.com/text-to-speech)
- [Azure Text-To-Speech](https://azure.microsoft.com/en-us/products/ai-services/text-to-speech)
- [Amazon Polly](https://aws.amazon.com/polly/)

The resulting dataset features roughly twice the samples per language (every `human` sample usually has its `synthetic` counterpart).


- **Curated by:** [KONTXT by RealNetworks](https://realnetworks.com/kontxt)
- **Funded by:** [RealNetworks](https://realnetworks.com/)
- **Language(s) (NLP):** English, German, Spanish, French, Italian, Dutch, Polish, Swedish
- **License:** [Apache 2.0](https://www.apache.org/licenses/LICENSE-2.0) for the code, [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/) for the dataset

### Dataset Sources

The original FLEURS dataset was downloaded from [HuggingFace](https://huggingface.co/datasets/google/fleurs).

- **FLEURS Repository:** [HuggingFace](https://huggingface.co/datasets/google/fleurs)
- **FLEURS Paper:** [arXiv](https://arxiv.org/abs/2205.12446)

- **Paper:** Synthetic speech detection with Wav2Vec 2.0 in various language settings

## Uses

This dataset is best used to train synthetic speech detection. Each sample contains an `Audio` feature, and a label: `human` or `synthetic`.

### Direct Use

The following snippet of code demonstrates loading the training split for English:

```python
from datasets import load_dataset

fleurs_hs = load_dataset(
    "realnetworks-kontxt/fleurs-hs",
    "en_us",
    split="train",
    trust_remote_code=True,
)
```

To load a different language, change `en_us` into one of the following:
- `de_de` for German
- `es_419` for Spanish
- `fr_fr` for French
- `it_it` for Italian
- `nl_nl` for Dutch
- `pl_pl` for Polish
- `sv_se` for Swedish

To load a different split, change the `split` value to `dev` or `test`.

The `trust_remote_code=True` parameter is necessary because this dataset uses a custom loader. To check out which code is being ran, check out the [loading script](./fleurs-hs.py).

## Dataset Structure

The dataset data is contained in the [data directory](https://huggingface.co/datasets/realnetworks-kontxt/fleurs-hs/tree/main/data).

There exists 1 directory per language.

Within those directories, there is a directory named `splits`; it contains 1 file per split:
- `train.tar.gz`
- `dev.tar.gz`
- `test.tar.gz`

Those `.tar.gz` files contain 2 directories:
- `human`
- `synthetic`

Each of these directories contain the `.wav` files for the label (and split). Keep in mind the the two directories can't be merged as they share most of their file names. An identical file name implies a speaker-voice pair, ex. `human/123.wav` and `synthetic/123.wav`.

Finally, back to the language directory, it contains 4 metadata files, which are not used in the loaded dataset, but might be useful to researchers:
- `recording-metadata.csv`
  - contains the transcript ID, file name, split and gender of the original FLEURS samples
- `recording-transcripts.csv`
  - contains the transcrpits of the original FLEURS samples
- `voice-distribution.csv`
  - contains the TTS vendor, TTS name, TTS engine, FLEURS gender and TTS gender for each ID-file name pair
  - useful for tracking what models were used to get specific synthetic samples
- `voice-metadata.csv`
  - contains the groupation of TTS' used alongside the splits they were used for

### Sample

A sample contains contains an Audio feature `audio`, and a string `label`.

```
{
  'audio': {
    'path': 'human/10004088536354799741.wav',
    'array': array([0., 0., 0., ..., 0., 0., 0.]),
    'sampling_rate': 16000
  },
  'label': 'human'
}
```

## Citation

The dataset is featured alongside our paper, **Synthetic speech detection with Wav2Vec 2.0 in various language settings**, which will be published on IEEE International Conference on Acoustics, Speech, and Signal Processing Workshops (ICASSPW). We'll provide links once it's available online.

**BibTeX:**

Note, the following BibTeX is incomplete - we'll update it once the actual one is known.

```
@inproceedings{dropuljic-ssdww2v2ivls
  author={Dropuljić, Branimir and Šuflaj, Miljenko and Jertec, Andrej and Obadić, Leo}
  booktitle={2024 IEEE International Conference on Acoustics, Speech, and Signal Processing Workshops (ICASSPW)}
  title={Synthetic speech detection with Wav2Vec 2.0 in various language settings}
  year={2024}
  volume={}
  number={}
  pages={1-5}
  keywords={Synthetic speech detection;text-to-speech;wav2vec 2.0;spoofing attack;multilingualism}
  doi={}
}
```

## Dataset Card Authors

- [Miljenko Šuflaj](https://huggingface.co/suflaj)

## Dataset Card Contact

- [Miljenko Šuflaj](mailto:msuflaj@realnetworks.com)