---
language:
- he
pretty_name: ILSpeech
viewer: false
tags:
- tts
- text-to-speech
- hebrew
task_categories:
- text-to-speech
---

# ILSpeech

Hebrew phonemes TTS Dataset 🗣️

## Dataset content

This dataset contains around 2 hours of studio-quality speech.

It includes two speakers (merged into a single dataset):

- Speaker 1: 40m male speaker (~578 samples)
- Speaker 2: 1h 30m male speaker (~951 samples)

It includes clean Hebrew text and IPA phonemes. The audio is in 22.05 kHz mono format with clean recordings.

## License

Non commercial. See [LICENSE](LICENSE)

## Sample

- Speaker 1:

<audio controls src="https://cdn-uploads.huggingface.co/production/uploads/63cfcf99e23b90128c66be2a/13JJXEgB5K08x5B47CsFG.wav"></audio>

- Speaker 2:

<audio controls src="https://cdn-uploads.huggingface.co/production/uploads/63cfcf99e23b90128c66be2a/DYElS9v--nTY5jlvQDduy.wav"></audio>

## Format

Annotations are provided in `metadata.csv`, `metadata_train.csv`, and `metadata_test.csv` (150 samples held-out test set).

Format: `id|ipa|text` — example: `speaker1_000|kotˈeʁet|כותרת` where `speaker1_000` points to `./wav/speaker1_000.wav`. Each wav is 2 to 25 seconds.

and the phonemization was done using https://github.com/thewh1teagle/phonikud, with manual corrections for accuracy.

## Training

You can find training receipes for the dataset in [github.com/thewh1teagle/ilspeech-train](https://github.com/thewh1teagle/ilspeech-train)

## Phonemize

The dataset comes with phonemes already so you don't need this for train, 

However you do need it after you train the model.

you can phonemize the text using [Phonikud](https://github.com/thewh1teagle/phonikud)