datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
IndicVoices
IndicVoices: Towards building an Inclusive Multilingual Speech Dataset for Indian Languages
Updates
[23 December 2025] We now have 11,200 hours of transcribed data! 🎉
Overview
INDICVOICES is a dataset of natural and spontaneous speech containing a total of 23.7K hours of read (8%), extempore (76%) and conversational (15%) audio from 51K speakers covering 400+ Indian districts and 22 languages. Of these 23.7K hours, 11.2K hours have… See the full description on the dataset page: https://huggingface.co/datasets/ai4bharat/IndicVoices.indicvoices_r
IndicVoices-R: Multilingual, Multi-Speaker Speech Corpus for Indian TTS
Dataset Summary
IndicVoices-R (IV-R) is the largest multilingual Indian text-to-speech (TTS) dataset derived from an automatic speech recognition (ASR) dataset. It contains 1,704 hours of high-quality speech from 10,496 speakers across 22 Indian languages. This dataset is designed to enhance the development of robust Indian TTS models by providing diverse speaker demographics, natural… See the full description on the dataset page: https://huggingface.co/datasets/ai4bharat/indicvoices_r.IndicSynth
IndicSynth: Indian Multilingual Audio Deepfake Detection & Anti-Spoofing Dataset
A Large-Scale Multilingual Synthetic Speech Dataset for Low-Resource Indian Languages to facilitate audio deepfake detection and anti-spoofing research
🏆 Outstanding Paper Award, ACL 2025
🧠 Overview
IndicSynth is a novel multilingual synthetic speech dataset designed to advance multilingual audio deepfake detection (ADD) and anti-spoofing research. It covers 12 low-resource Indian… See the full description on the dataset page: https://huggingface.co/datasets/vdivyasharma/IndicSynth.indic-dialect-asr
Indic Dialect ASR Dataset
A multilingual ASR dataset covering 30 Indic dialect/languages with 2.8M+ samples.
Usage
from datasets import load_dataset
# Load a specific language
ds = load_dataset("grushaaaaa/indic-dialect-asr", "assamese", split="train")
Features
audio: 16kHz WAV audio
sentence: Transcription text
language: Language name
source: Source dataset
IndicSynth
IndicSynth: Indian Multilingual Audio Deepfake Detection & Anti-Spoofing Dataset
A Large-Scale Multilingual Synthetic Speech Dataset for Low-Resource Indian Languages to facilitate audio deepfake detection and anti-spoofing research
🏆 Outstanding Paper Award, ACL 2025
🧠 Overview
IndicSynth is a novel multilingual synthetic speech dataset designed to advance multilingual audio deepfake detection (ADD) and anti-spoofing research. It covers 12 low-resource Indian… See the full description on the dataset page: https://huggingface.co/datasets/ksmashhero/IndicSynth.indic-tts-966h
Indic-TTS-966h
Six-language Indian TTS corpus: ~966 hours of paired speech and text, 24 kHz mono WAV
clips with sentence-level transcripts in native scripts (natural English code-switching
preserved).
Subset
Clips
Hours
bengali
18,343
94.9
malayalam
30,548
192.5
marathi
34,327
213.4
punjabi
28,083
161.8
tamil
26,817
171.1
telugu
21,923
132.8
Columns: audio (24 kHz mono), file_name, transcript. One config per language:
from datasets import load_dataset… See the full description on the dataset page: https://huggingface.co/datasets/psk/indic-tts-966h.quran-terjemahan-indonesia-audio-by-ayah
Audio Terjemahan Al-Qur'an Indonesia per Ayat
Dataset ini berisi audio terjemahan Al-Qur'an bahasa Indonesia yang dipisahkan per ayat.
Audio dibuat menggunakan Gemini TTS dari teks terjemahan Al-Qur'an bahasa Indonesia. Setiap file audio mewakili satu ayat.
Isi Dataset
6.236 file audio WAV
6.236 file audio M4A/AAC terkompresi untuk aplikasi mobile
Bahasa Indonesia
Satu file audio untuk setiap ayat
Format WAV dan M4A/AAC
Disusun berdasarkan nomor surah dan nomor… See the full description on the dataset page: https://huggingface.co/datasets/insanalamin/quran-terjemahan-indonesia-audio-by-ayah.indic-diarbench
Indic DiarBench
A multilingual joint diarization and ASR benchmark for Indian languages, spanning all 22 scheduled languages of India with approximately 108 hours of natural multi-speaker audio.
Paper: Indic DiarBench: A Multilingual Joint Diarization and ASR Benchmark for Indian Languages (Interspeech 2026)
Dataset Summary
Indic DiarBench is a conversational speech benchmark designed to evaluate speaker-attributed ASR in realistic multi-speaker settings for… See the full description on the dataset page: https://huggingface.co/datasets/sarvamai/indic-diarbench.IndicTTS-Deepfake-Challenge-Data
IndicTTS Deepfake Detection Challenge
Participants will use the SherryT997/IndicTTS-Deepfake-Challenge-Data dataset, hosted on Hugging Face. This dataset consists of train and test splits and contains speech samples in 16 Indian languages, along with metadata for each audio clip.
🚀 Dataset to Use: SherryT997/IndicTTS-Deepfake-Challenge-Data
This is the official dataset for the challenge and must be used for training and evaluation.
📌 Dataset Structure… See the full description on the dataset page: https://huggingface.co/datasets/SherryT997/IndicTTS-Deepfake-Challenge-Data.IndicTTS-Hindi
Hindi Indic TTS Dataset
This dataset is derived from the Indic TTS Database project, specifically using the Hindi monolingual recordings from both male and female speakers. The dataset contains high-quality speech recordings with corresponding text transcriptions, making it suitable for text-to-speech (TTS) research and development.
Dataset Details
Language: Hindi
Total Duration: ~10.33 hours (Male: 5.16 hours, Female: 5.18 hours)
Audio Format: WAV
Sampling Rate: 48000Hz… See the full description on the dataset page: https://huggingface.co/datasets/SPRINGLab/IndicTTS-Hindi.IndicTTS-EnglishIndicSynth
IndicSynth: Indian Multilingual Audio Deepfake Detection & Anti-Spoofing Dataset
A Large-Scale Multilingual Synthetic Speech Dataset for Low-Resource Indian Languages to facilitate audio deepfake detection and anti-spoofing research
🏆 Outstanding Paper Award, ACL 2025
🧠 Overview
IndicSynth is a novel multilingual synthetic speech dataset designed to advance multilingual audio deepfake detection (ADD) and anti-spoofing research. It covers 12 low-resource Indian… See the full description on the dataset page: https://huggingface.co/datasets/mrunmai18/IndicSynth.Indic-total-New-TTS-Merge
Indic Total TTS Merge
Merged TTS dataset with 13 Indic languages. All audio clips are >= 3.0 seconds duration.
Languages
assamese, bengali, english, gujarati, hindi, kannada, malayalam, marathi, nepali, odia, punjabi, tamil, telugu
Columns
audio: Audio data
text: Transcript text
duration: Duration in seconds (all >= 3.0s)
language: Language name
indic_monovoiceIndicDiarBench-speaker-retrieval
Indic DiarBench speaker retrieval (MTEB)
Indic DiarBench reshaped for speaker retrieval across the 22 scheduled languages
of India: given a clip of one speaker, find other clips of that same speaker.
Source: sarvamai/indic-diarbench at revision 92877ba, cc-by-4.0, official
test split. Turns are cut by their annotated times, restricted to 2 to 15
seconds, and turns overlapping a different speaker are dropped. identity pairs
the recording session with the speaker, because speaker… See the full description on the dataset page: https://huggingface.co/datasets/vnahata/IndicDiarBench-speaker-retrieval.tts_indicindian-english-nptel-v0indic-superb-whisperindic-Malayalam-PDIndicTTS_Bengali
Bengali Indic TTS Dataset
This dataset is derived from the Indic TTS Database project, specifically using the Bengali monolingual recordings from both male and female speakers. The dataset contains high-quality speech recordings with corresponding text transcriptions, making it suitable for text-to-speech (TTS) research and development.
Dataset Details
Language: Bengali
Total Duration: ~15.06 hours (Male: 10.05 hours, Female: 5.01 hours)
Audio Format: WAV
Sampling Rate:… See the full description on the dataset page: https://huggingface.co/datasets/SPRINGLab/IndicTTS_Bengali.indicV-cleanedIndicTTS_Gujarati
task_categories:
- text-to-speech
language:
- gj
pretty_name: Gujarati Indic TTS dataset
size_categories:
- n<1K
Gujarati Indic TTS Dataset
This dataset is derived from the Indic TTS Database project, specifically using the Gujarati monolingual recordings from both male and female speakers. The dataset contains high-quality speech recordings with corresponding text transcriptions, making it suitable for text-to-speech (TTS) research and development.
Gujarati… See the full description on the dataset page: https://huggingface.co/datasets/SPRINGLab/IndicTTS_Gujarati.IndicVoices-R_Hindiindic-voices-hinglish-nospeakeroverlap-spon3.3-acronyms-fixed2IndicTTS_Telugu
Telugu Indic TTS Dataset
This dataset is derived from the Indic TTS Database project, specifically using the Telugu monolingual recordings from both male and female speakers. The dataset contains high-quality speech recordings with corresponding text transcriptions, making it suitable for text-to-speech (TTS) research and development.
Dataset Details
Language: Telugu
Total Duration: ~8.74 hours (Male: 4.47 hours, Female: 4.27 hours)
Audio Format: WAV
Sampling Rate:… See the full description on the dataset page: https://huggingface.co/datasets/SPRINGLab/IndicTTS_Telugu.IndicTTS_Tamil
Tamil Indic TTS Dataset
This dataset is derived from the Indic TTS Database project, specifically using the Tamil monolingual recordings from both male and female speakers. The dataset contains high-quality speech recordings with corresponding text transcriptions, making it suitable for text-to-speech (TTS) research and development.
Dataset Details
Language: Tamil
Total Duration: ~20.33 hours (Male: 10.3 hours, Female: 10.03 hours)
Audio Format: WAV
Sampling Rate:… See the full description on the dataset page: https://huggingface.co/datasets/SPRINGLab/IndicTTS_Tamil.indonesian-voice-transcription-1.4.9a-rtts-indo
agufsamudra/tts-indo
agufsamudra/tts-indo is a preprocessed Indonesian speech dataset designed for training Text-to-Speech (TTS) models. This dataset is derived from the original Dataset TTS Indo available on Kaggle.
Dataset Details
Number of Examples: 114,036
Dataset Size: ~4GB
Audio Sampling Rate: 16,000 Hz
Features:
audio: WAV audio recordings
text: Transcription of the audio
Data Structure
Each sample in the dataset contains:
audio: A dictionary… See the full description on the dataset page: https://huggingface.co/datasets/agufsamudra/tts-indo.synthetic-speech-indicindic-audio
indic-audio
A multi-speaker synthetic speech dataset for Hindi, Indian English, and Hinglish (Hindi in Latin script): 5,160 clips, 7.1 hours, 15 voices. Built to train goonj-1-82M, an edge-sized Indian-language TTS model.
Summary
Clips
5,160
Total audio
7.13 h (avg 5.0 s/clip)
Voices
15 (14 named personas + 1 Hindi "language bed")
Languages
Hindi (Devanagari), Indian English, Hinglish (romanized)
Format
44.1 kHz; WAV (bed_hindi) and MP3… See the full description on the dataset page: https://huggingface.co/datasets/BH-Builds/indic-audio.