datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
TTS-Hungarian
TTS-Hungarian
A large-scale, high-quality Hungarian speech dataset for text-to-speech and automatic speech recognition.
Data Source
Derived from MEK (Magyar Elektronikus Könyvtár) — Hungarian audiobooks.
Dataset Statistics
Metric
Value
Total samples
253,116
Total duration
702 hours
Unique speakers
100
Average duration
10.0 seconds
Average DNSMOS
3.68
Features
Field
Type
Description
__key__
string
Unique sample… See the full description on the dataset page: https://huggingface.co/datasets/datadriven-company/TTS-Hungarian.Hungarian-Speech-Dataset
🎧 Hungarian Speech Dataset
The Hungarian Speech Dataset is a high-quality speech audio dataset designed to support advanced AI systems that depend on diverse audio data and reliable voice data for multilingual model training. It comprises 169 hours of recordings across 743 files, provided in MP3 and WAV formats, with a total size of 134 MB. This structured audio dataset ensures balanced speaker representation, featuring 46% female and 54% male speakers, and an age distribution… See the full description on the dataset page: https://huggingface.co/datasets/Speech-data/Hungarian-Speech-Dataset.YodaLingua-Hungarian
YodaLingua-Hungarian
YodaLingua is a high-quality speech dataset designed for training text-to-speech (TTS) systems, ASR models, and any application requiring clean, well-aligned audio–text pairs.This release contains the Hungarian portion of the multilingual YodaLingua collection.
🧾 Dataset Overview
Property
Value
Total clips
80,740 audio–transcription pairs
Total duration
206 hours
Speakers
1,856 distinct speakers
Audio format
MP3 • mono • 24 kHz •… See the full description on the dataset page: https://huggingface.co/datasets/Thomcles/YodaLingua-Hungarian.