datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
synthetic_dem
Dataset Card for synthetic_dem
Dataset Summary
The Synthetic DEM Corpus is the result of the first phase of a collaboration between El Colegio de México (COLMEX) and the Barcelona Supercomputing Center (BSC).
It all began when COLMEX was looking for a way to have its Diccionario del Español de México (DEM), which can be accessed online, include the option to play each of its words with a Mexican accent through synthetic speech files. On the other hand, BSC is always on… See the full description on the dataset page: https://huggingface.co/datasets/projecte-aina/synthetic_dem.SynthGT
SynthGT
A Synthetic Solo-Singing Dataset for Singing-Oriented Forced Alignment
Authors
Silas Antonisen, Iván López-Espejo
Associated paper submitted to IEEE Transactions on Audio, Speech and Language Processing.
Overview
SynthGT (Synthetic Ground Truth) is a synthetic English solo-singing dataset containing 4,900 singing performances with automatically generated phoneme boundary annotations.
The dataset was created through music… See the full description on the dataset page: https://huggingface.co/datasets/Silasimo/SynthGT.icelandic_asr
Icelandic ASR Collection
This repository collects six Icelandic speech corpora in directly loadable
Parquet form. Audio is embedded as 16 kHz mono FLAC bytes. The repository is a
convenience repackaging: the linked CLARIN-IS records and original dataset
repositories remain the canonical sources and should be cited when using the
data.
No configuration is selected by default. Choose a corpus configuration and,
for this large collection, normally choose a split explicitly.… See the full description on the dataset page: https://huggingface.co/datasets/Aalto-Speech-Synthesis/icelandic_asr.barranquenho-ipa-dict-synthetic
Barranquenho IPA Pronunciation Dictionary
The first and only IPA pronunciation dictionary of Barranquenho — the
Ibero-Romance contact variety spoken in Barrancos (Baixo Alentejo, Portugal),
a mixed system born of centuries of Portuguese–Spanish (Extremaduran /
Andalusian) contact on the raia. Every headword is written in the
Convenção Ortográfica do Barranquenho (2025) orthography and paired with a
broad-phonemic IPA transcription plus Portuguese and Spanish glosses.
This… See the full description on the dataset page: https://huggingface.co/datasets/TigreGotico/barranquenho-ipa-dict-synthetic.multivoice-synthetic-speech
Synthetic Voice Samples · Africa
Synthetic speech. No human speaker was recorded for any clip in this dataset.
Generated with afrispeech-synth: text from
africa-corpus, normalised to a
universal orthography with africa-g2p, spoken by
Google Gemini's Live API.
17,010 clips · 38.8 hours · 566 languages · 30 voices
Every clip is a distinct sentence — no sentence is repeated
Each language is read by up to 30 different voices, one sentence per voice
~1.29 hours per voice… See the full description on the dataset page: https://huggingface.co/datasets/AfriSpeech/multivoice-synthetic-speech.synthetic_transcript_pt
Portuguese Speech Dataset with Multiple Training Configurations
A comprehensive Portuguese speech dataset offering three distinct training configurations for speech recognition research, each designed for different experimental scenarios and training paradigms.
🎯 Dataset Configurations Overview
This dataset provides three carefully curated subsets to enable comprehensive speech recognition research:
Configuration
Training Data
Validation
Test
Total Samples
Use Case… See the full description on the dataset page: https://huggingface.co/datasets/yuriyvnv/synthetic_transcript_pt.masri_synthetic
Dataset Card for masri_synthetic
Dataset Summary
The MASRI-SYNTHETIC is a corpus made out of synthesized speech in Maltese. The text-to-speech (TTS) system utilized to produce the utterances was developed by the Research & Development Department of Crimsonwing p.l.c.
The sentences used to create the corpus were extracted from the MLRS Corpus, which is a corpus of written or transcribed Maltese divided into different genres, including: culture, news, academic, religion… See the full description on the dataset page: https://huggingface.co/datasets/MLRS/masri_synthetic.SIWIS_French_Speech_Synthesis_Database
SIWIS French Speech Synthesis Database
This README provides a concise description of the dataset, including its structure, file naming conventions, and known labeling issues. Additionally, suggestions for potential improvements are outlined in the TODO section.
The dataset is distributed under the Creative Commons Attribution 4.0 International (CC BY 4.0) license, permitting its use for any purpose.
For more details about the database design and recording process, please refer… See the full description on the dataset page: https://huggingface.co/datasets/Aviv-anthonnyolime/SIWIS_French_Speech_Synthesis_Database.Aslema-Synth-TN
Aslema-Synth-TN
Aslema-Synth-TN is a fully synthetic Tunisian Derja corpus for spoken language
understanding: speech annotated for intent and for slot filling. It was built for
NADI 2026 Shared Task 5 to cover the intents and slots that
are rare or absent in the real SLURP-TN training split, and it is the augmentation
set behind the Aslema system, which ranked 1st in slot filling on the official test
set.
No human was recorded for this dataset. An LLM wrote the utterance text, a… See the full description on the dataset page: https://huggingface.co/datasets/QCRI/Aslema-Synth-TN.meddies-asr-synth-dialog
Meddies ASR — Synthetic Dialog Speech (vi/en/zh)
Synthetic doctor–patient consultations for ASR training, generated with Fish Audio TTS
(s2.1-pro-free, 16 kHz mono FLAC) over curated conversational voice registries
(en 75 / zh 165 voices) with 22 compositional scenario profiles driving expressiveness
(disfluencies, emotion palettes, speed jitter, age-band casting, pause pacing).
Target: 1,000 h per language.
Configs
vi_dialog: Vietnamese clinical dialogue turns… See the full description on the dataset page: https://huggingface.co/datasets/Meddies/meddies-asr-synth-dialog.Synthetic-Medical-Speech-Dataset
Synthetic Medical Speech Dataset
Overview
Synthetic Medical Speech Dataset is a synthetic dataset of audio–text pairs designed for developing and evaluating automatic speech recognition (ASR) models in the medical domain.The corpus contains thousands of short audio clips generated from medically relevant text using a text-to-speech (TTS) system.Each clip is paired with its corresponding transcript.Because all content is synthetically produced, the dataset does not contain… See the full description on the dataset page: https://huggingface.co/datasets/Hani89/Synthetic-Medical-Speech-Dataset.synthetic_speech_commands_PA_taggedsynthetic-tts
Munni — Hindi/English Synthetic TTS Voice Dataset
Single-speaker, code-mixed Hindi (Devanagari) + English speech dataset built for
XTTS-v2 fine-tuning. Domain is call-center / customer-support style dialogue
(warranty, billing, appointment scheduling), with scripted placeholder phone
numbers spoken digit-by-digit.
8,796 clips / 18.16 hours, 24 kHz mono 16-bit PCM
Split: train 8,621 / validation 175 (98/2, seed 42)
Duration per clip: mean 7.43s, min 3.18s, max 11.00s
Single… See the full description on the dataset page: https://huggingface.co/datasets/routsourav1729/synthetic-tts.synthetic-parallel-external
Synthetic Parallel EN↔LG — external
Voice-controlled synthetic parallel speech dataset for Luganda-English
speech-to-speech translation, generated by the Hibiki-Zero fine-tuning pipeline.
Generation
Component
Model
Translation
Sunbird/translate-nllb-3.3b-salt
TTS
Sunbird/orpheus-3b-tts-multilingual
English speakers: salt_eng_0001, salt_eng_0002, salt_eng_0003
Luganda speakers: salt_lug_0001, waxal_lug_0001, waxal_lug_0002, waxal_lug_0003, waxal_lug_0004… See the full description on the dataset page: https://huggingface.co/datasets/yigagilbert/synthetic-parallel-external.ghana-twi-synthesized-speech
Ghana Twi & Code-Switching Synthesized Speech Dataset
Synthesized text-to-speech audio for Twi and English-Twi code-switching sentences.
How this dataset was built
1. Source text
The sentences come from two sources, combined and deduplicated:
A sentence-subset sampled from
ghananlpcommunity/pristine-twi-english
(greedy set-cover over 3,000 articles so that every word appearing in the corpus is present in at
least one selected sentence).
The… See the full description on the dataset page: https://huggingface.co/datasets/ghanaopenai/ghana-twi-synthesized-speech.synthetic_transcript_nl
Dutch Synthetic Speech Transcripts
This dataset contains 34,898 synthetic Dutch speech samples generated using GPT-4o-mini for transcript creation and OpenAI's TTS-1 model for speech synthesis. It was designed to augment Automatic Speech Recognition (ASR) training for low-resource scenarios, matching the linguistic distribution of Common Voice 17.0 Dutch.
Dataset Description
Purpose
This dataset addresses the challenge of limited labeled speech data for Dutch… See the full description on the dataset page: https://huggingface.co/datasets/yuriyvnv/synthetic_transcript_nl.korean-full-duplex-synthetic-dataset-preview
Korean Full-Duplex Synthetic Dataset Preview
Overview
Public preview of a Korean full-duplex synthetic speech dataset. This
repository contains 100 conversations sampled from a corpus of 89,273
conversations (2,000.5 hours); it does not publish the full corpus audio.
Preview contents
100 conversation WAV files
data/representative.jsonl
24 kHz, mono, 16-bit PCM
Events: normal, barge_in, backchannel, cutoff_by_user
Annotation format… See the full description on the dataset page: https://huggingface.co/datasets/Wi-Fi/korean-full-duplex-synthetic-dataset-preview.Hausa-Synthetic-ASR-Dataset-XTTSSynthetic Hausa ASR dataset generated using a fine-tuned version of the XTTS-v2 model.
Sample rate: 24kHz.
Total duration: 574 hours.
synthetic_maritime_radio_communication
MARTTS: Maritime Radio Text-To-Speech Synthetic Corpus
Synthetic VHF Maritime Communication Data for Robust ASR Evaluation
Dataset accompanying the paper:A Text-to-Speech Framework for Generating Synthetic Maritime Radio Communications in ASR Evaluation
Dataset Summary
MARTTS is an open-source synthetic speech corpus designed to evaluate and stress-test Automatic Speech Recognition (ASR) systems operating in maritime VHF radiotelephony environments.
The… See the full description on the dataset page: https://huggingface.co/datasets/bonalor/synthetic_maritime_radio_communication.arabic-english-code-switching-synthetic-asr
Synthetic Arabic-English Code-Switched Speech for ASR
This dataset contains synthetic speech generated for Egyptian Arabic-English code-switched automatic speech recognition. It is published separately from the human review annotations so the human audio remains in its upstream Hugging Face repository.
Configurations
Configuration
Train
Test
Publication status
synthetic
8,655
962
Contains 5,673 ArE-CSTD-derived texts; noncommercial/share-alike terms… See the full description on the dataset page: https://huggingface.co/datasets/abdo1819/arabic-english-code-switching-synthetic-asr.nepali-tts-synthetic-v2
Nepali TTS Synthetic v2
383,298 synthetic Nepali (ne) speech/text pairs, 24 kHz mono 16-bit WAV embedded
as-is (no re-encode, no resampling).
Generated by the synthetic_pipeline in milanakdj/TTS_training: Edge TTS
synthesis → optional voice conversion against a pool of 600 real multi-speaker
reference clips → ASR-based QC gate on character error rate.
Read this before training on it
Only 48% of rows are voice-converted. Each row carries a kept field
recording… See the full description on the dataset page: https://huggingface.co/datasets/milanakdj/nepali-tts-synthetic-v2.burmese-synthetic-speech-corpus
Burmese Synthetic Speech Corpus (DatarrX/burmese-synthetic-speech-corpus)
Overview
The Burmese Synthetic Speech Corpus is a high-fidelity, manually curated audio dataset specifically designed to advance Text-to-Speech (TTS) systems, speech recognition, and other audio-driven Machine Learning tasks for the Burmese (Myanmar) language.
Created by DatarrX, this dataset bridges the gap in low-resource speech technologies by providing highly natural, native-sounding… See the full description on the dataset page: https://huggingface.co/datasets/DatarrX/burmese-synthetic-speech-corpus.turkish-synthetic-whisper-rounds1-4.5-355h
Turkish Synthetic Whisper Rounds 1–4.5
Archival release of the exact 199,590-record, 355.186-hour synthetic corpus used to fine-tune the final Round 4.5 Whisper Tiny and Base models.
Each row in train.jsonl references both:
training_audio: the exact clean or exactly-once postprocessed waveform used in training; and
clean_audio: its original synthetic clean waveform.
Audio is SHA-256 deduplicated and stored in deterministic tar.zst shards. Common Voice/FLEURS evaluation audio… See the full description on the dataset page: https://huggingface.co/datasets/erayyapagci/turkish-synthetic-whisper-rounds1-4.5-355h.synthetic-multilingual-speaker-diarization
Synthetic Multilingual Speaker Diarization Dataset
This dataset contains synthetic multilingual speaker diarization data with Hindi, English, and Punjabi audio samples.
Dataset Structure
├── audio/ # WAV audio files (16kHz) - 3417 files
├── all_samples_combined.csv # Complete dataset annotations (with silence)
└── all_visible_combined.csv # Visible dataset annotations (without silence)
Statistics
Total samples: 3417 audio… See the full description on the dataset page: https://huggingface.co/datasets/noty7gian/synthetic-multilingual-speaker-diarization.synthetic-speaker-diarization-dataset-fa-large-3000synthetic-speech-diarization-ru
synthetic-speech-diarization-ru
Synthetic speech diarization dataset in Parquet format.
Dataset Details
Number of tracks: 2000
Sampling rate: 16000 Hz
Audio format: Embedded in Parquet files (Audio feature compatible)
Storage: Parquet format for efficient loading
Dataset Structure
The dataset contains audio tracks with speaker diarization annotations, stored directly in Parquet format.
Features
audio: Audio waveform (Audio feature with array and… See the full description on the dataset page: https://huggingface.co/datasets/ivkond/synthetic-speech-diarization-ru.romanian_speech_synthesis_0_8_1\
The Romanian speech synthesis (RSS) corpus was recorded in a hemianechoic chamber (anechoic walls and ceiling; floor partially anechoic) at the University of Edinburgh. We used three high quality studio microphones: a Neumann u89i (large diaphragm condenser), a Sennheiser MKH 800 (small diaphragm condenser with very wide bandwidth) and a DPA 4035 (headset-mounted condenser). Although the current release includes only speech data recorded via Sennheiser MKH 800, we may release speech data recorded via other microphones in the future. All recordings were made at 96 kHz sampling frequency and 24 bits per sample, then downsampled to 48 kHz sampling frequency. For recording, downsampling and bit rate conversion, we used ProTools HD hardware and software. We conducted 8 sessions over the course of a month, recording about 500 sentences in each session. At the start of each session, the speaker listened to a previously recorded sample, in order to attain a similar voice quality and intonation.stortinget_speech_corpus_v1.0
Dataset Card for Stortinget Speech Corpus V1.0
Overview
This is the WebDataset version of the Stortinget Speech Corpus V1.0, originally created by the National Library of Norway. We re-organize it into WebDataset format for better usability.
The Stortinget Speech Corpus (SSC) is a 5000+ hours speech dataset for weak supervision ASR created from audio andaligned proceedings text from Stortinget, the Norwegian Parliament. For more information, please refer to the original… See the full description on the dataset page: https://huggingface.co/datasets/Aalto-Speech-Synthesis/stortinget_speech_corpus_v1.0.Synthetic_Turkish_TTS_Data
Synthetic Turkish TTS Data
This dataset was created by generating synthetic Turkish text across multiple speech scenarios. The text was produced in the following domains: finance_master, cs_master, parcel_delivery, ecommerce, telecom, isp_support, technical_support, subscription, insurance, health_appointments, public_services, education_registration, and daily_speech.
These synthetic texts were then synthesized with a high-quality Turkish TTS model. The dataset is intended to be… See the full description on the dataset page: https://huggingface.co/datasets/Anilosan15/Synthetic_Turkish_TTS_Data.multilingual-synthetic-tts
Multilingual Synthetic TTS Dataset
🏆 Submitted to the Uncharted Data Challenge
hosted by Adaption Labs — credit to
Adaptive Data by Adaption for organizing the hackathon.
A large-scale synthetic multilingual speech dataset — 68,677 clips across
9 languages, generated with Qwen3-TTS-12Hz-1.7B-Base
using zero-shot voice cloning from 5 reference speakers.
Intended for training and evaluating TTS, ASR, voice conversion, and
multilingual speech models. Each clip is paired with the… See the full description on the dataset page: https://huggingface.co/datasets/Reubencf/multilingual-synthetic-tts.