Datasets | Mozilla Data Collective

Community

Urdu Multi-Speaker TTS Dataset

An Urdu multi-speaker TTS dataset distributed in 36 zip files, each containing audio files and a TSV mapping file, with approximately 10 hours of speech.

License: CC-BY-NC-4.0

Locale: urd

Task: TTS

Format: WEBM, TSV

Size: 514.54 MB

MirasAI

Punjabi 10 Hours TTS

10-hour Punjabi TTS dataset in Shahmukhi script with paired audio and transcripts, designed for speech synthesis and Punjabi language technology.

License: CC-BY-NC-SA-4.0

Locale: pnb

Task: TTS

Format: WEBM, TSV

Size: 481.96 MB

MDC DatasetsCompensated

Single Speaker Urdu TTS dataset (Female Speaker, South Punjab)

A high-quality single-speaker Urdu text-to-speech (TTS) dataset with 25 hours of paired speech recordings and transcriptions for speech synthesis.

License: MDC DLA-1.0

Locale: ur

Task: TTS

Format: WAV, TSV

Size: 6.02 GB

PROXIMA

Synthetic Urdu 2

Synthetic Urdu sentences audio dataset of 85,327 verse style audio-text pairs.

License: CC-BY-NC-SA-4.0

Locale: urd

Task: TTS

Format: Parquet

Size: 14.80 GB

MirasAI

Saraiki 10 Hours TTS Dataset

A 10-hour Saraiki text-to-speech dataset consisting of recorded speech and aligned transcripts, designed for speech synthesis research and development.

License: CC-BY-NC-SA-4.0

Locale: srk

Task: TTS

Format: WEBM, TSV

Size: 584.44 MB

MirasAICompensated

🚧 Noakhali 10 Hours TTS (Male Speaker) 🚧

A high-quality Noakhali TTS dataset with paired speech recordings and text transcriptions for speech synthesis, linguistic research, and language preservation.

License: MirasAI Data License

Locale: oak

Task: TTS

Format: WAV, TSV

Size: 2.87 GB

MDC Community Concierge

VoxForge - Persian

16 minutes (157 utterances) of read speech of ‭فارسی‬ (Persian).

License: GPL-3.0

Locale: fa

Task: ASR

Format: WAV, TSV

Size: 58.20 MB

Common Voice

Common Voice Scripted Speech 26.0 - Urdu

A collection of read speech recordings in Urdu.

License: CC0-1.0

Locale: ur

Task: ASR

Format: MP3

Size: 5.79 GB

MirasAI

Associated Press Pakistan 74.4 Million Urdu Corpus

A 74.4M-word, 267,887-article Urdu news corpus from Associated Press Pakistan (2015-2026), released for non-commercial research with APP permission.

License: CC-BY-NC-SA-4.0

Locale: urd

Task: NLP

Format: TXT

Size: 143.39 MB

The University of Melbourne

Central Kurdish TTS dataset 1.0

This dataset contains high-quality single-speaker audio recordings in Central Kurdish (ckb), intended for building Text-to-Speech (TTS) and Automatic Speech recognition (ASR) systems. The dataset comprises 2 hours and 18 minutes of aligned audio and text data.

License: CC-BY-4.0

Locale: ckb

Task: TTS

Format: wav

Size: 293.45 MB

PROXIMA

Syntheic Urdu Audios

Synthetic Urdu conversational speech dataset of 7,963 audio text pairs featuring casual, humorous dialogue exchanges.

License: CC-BY-NC-SA-4.0

Locale: urd

Task: TTS

Format: PARQUET

Size: 4.58 GB

The University of Melbourne

Hawrami Kurdish TTS dataset 1.0

This dataset contains high-quality single-speaker audio recordings in Hawrami Kurdish (Hewrami, ISO 639-3:hac), also known as the Gorani language, intended for building Text-to-Speech (TTS) and Automatic Speech recognition (ASR) systems. The dataset comprises 5 hours and 15 minutes of aligned audio and text data. Hawrami is classified as Definitely Endangered by UNESCO.

License: CC-BY-4.0

Locale: hac

Task: TTS

Format: WAV

Size: 706.11 MB

Common Voice

Common Voice Scripted Speech 26.0 - Indus Kohistani

A collection of read speech recordings in Indus Kohistani.

License: CC0-1.0

Locale: mvy

Task: ASR

Format: MP3

Size: 448.96 MB

Weekly Kaleem Magazine Multan

Kaleem Magazine Urdu Corpus

This corpus is a collection of around 1.4 million tokens of Urdu language. The data was extracted from the archives of a famous Urdu magazine "Kaleem" published weekly from last 30 years. This corpus contains work of literature including stories, short stories, news, poetry, literary reports, fiction, non-fiction, and travelogues. The data is being shared with the approval of the authors. It aims to support linguistic research, language technology development, and cultural preservation.

License: CC-BY-NC-4.0

Locale: urd

Task: NLP

Format: TXT

Size: 2.74 MB

Kaleem Art Press

Kaleem Art Press Urdu Literature Corpus

This corpus is a collection of 1.44 million tokens of Urdu language . The data was produced under the Kaleem Art Press over the last fifteen years . The corpus contains work of literature including Stories, Short Stories, Novels, fiction, non-fiction, Travelogues, Poetry, Biography, and History. The data is being shared with the approval of the authors. It aims to support linguistic research, language technology development, and cultural preservation.

License: CC-BY-NC-4.0

Locale: ur

Task: OTH

Format: TXT

Size: 2.85 MB

Rana Printers Multan

Rana Printers Urdu Literature Corpus

This corpus comprises 1.68 million tokens of high-quality Urdu text collected over the past decade through Rana Printers. It includes a diverse range of literary genres such as stories, short stories, novels, fiction, non-fiction, poetry, and historical works. All content is shared with the authors’ approval. The dataset is intended to support linguistic research, Urdu language technology development, and the preservation of literary and cultural heritage.

License: CC-BY-NC-4.0

Locale: ur

Task: OTH

Format: TXT

Size: 3.00 MB

Common Voice

Common Voice Scripted Speech 26.0 - Khetrani

A collection of read speech recordings in Khetrani.

License: CC0-1.0

Locale: xhe

Task: ASR

Format: MP3

Size: 196.18 MB

MDC DatasetsCompensated

🚧 Southeast Asian Tamil TTS Dataset 🚧

2 hours of studio-quality, deliberate, read speech and corresponding text from a single speaker of Southeast Asian Tamil.

License: MDC DLA-1.0

Locale: ta

Task: TTS

Format: WAV, TSV

Size: 193 B

Taruen

Chuvash TTS

A ~5-hour speech dataset for Chuvash Text-to-Speech (TTS) research, featuring a single female speaker reading news and digits at a rapid tempo.

License: CC-BY-SA-4.0

Locale: cv

Task: TTS

Format: PARQUET

Size: 854.02 MB

Open Home Foundation

Chitwan 1.0

Text to speech dataset for Nepali, male speaker, approximately 1 hour of read speech.

License: CC0-1.0

Locale: ne-NE

Task: TTS

Format: WEBM

Size: 61.68 MB

Common Voice

Common Voice Scripted Speech 26.0 - Gujari

A collection of read speech recordings in Gujari.

License: CC0-1.0

Locale: gju

Task: ASR

Format: MP3

Size: 210.21 MB

Common Voice

Common Voice Scripted Speech 26.0 - Northern Hindko

A collection of read speech recordings in Northern Hindko.

License: CC0-1.0

Locale: hno

Task: ASR

Format: MP3

Size: 206.29 MB

Common Voice

Common Voice Spontaneous Speech 4.0 - Pashto

A collection of spontaneous responses to questions in Pashto.

License: CC0-1.0

Locale: ps

Task: ASR

Format: MP3

Size: 223.81 MB

MirasAI

Tamil Time Aligned Speech Dataset

5-hour Tamil speech dataset with time-aligned transcripts, designed for ASR, forced alignment, subtitle generation, and speech-language research.

License: CC-BY-NC-SA-4.0

Locale: tam

Task: ASR

Format: OGG, SRT

Size: 37.11 MB

All Datasets

Not sure where to start?

Urdu Multi-Speaker TTS Dataset

Punjabi 10 Hours TTS

Single Speaker Urdu TTS dataset (Female Speaker, South Punjab)

Synthetic Urdu 2

Saraiki 10 Hours TTS Dataset

🚧 Noakhali 10 Hours TTS (Male Speaker) 🚧

VoxForge - Persian

Common Voice Scripted Speech 26.0 - Urdu

Associated Press Pakistan 74.4 Million Urdu Corpus

Central Kurdish TTS dataset 1.0

Syntheic Urdu Audios

Hawrami Kurdish TTS dataset 1.0

Common Voice Scripted Speech 26.0 - Indus Kohistani

Kaleem Magazine Urdu Corpus

Kaleem Art Press Urdu Literature Corpus

Rana Printers Urdu Literature Corpus

Common Voice Scripted Speech 26.0 - Khetrani

🚧 Southeast Asian Tamil TTS Dataset 🚧

Chuvash TTS

Chitwan 1.0

Common Voice Scripted Speech 26.0 - Gujari

Common Voice Scripted Speech 26.0 - Northern Hindko

Common Voice Spontaneous Speech 4.0 - Pashto

Tamil Time Aligned Speech Dataset