ARTPARK-IISc/Vaani-transcription-part

This dataset is part of the Vaani dataset and consists of only transcribed speech data. It has a total duration of 788 hours, covering 30 languages. This table represents the audio and transcription duration data for various languages.

Language	Angika	Awadhi	Bagri	Bajjika	Bearybashe	Bengali	Bhatri	Bhili	Bhojpuri	Bundeli	Chhattisgarhi	Garhwali	Gondi	Gujarati	Halbi	Hindi	Kannada	Khariboli	Khortha	Konkani	Kumaoni	Kurmali	Kurukh	Magahi	Maithili	Malvani	Marathi	Marwari	Rajasthani	Sadri	Santali	Shekhawati	Surgujia	Surjapuri	Tamil	Telugu	Tulu	Urdu
Trans. Duration(Hrs)	3.53	0.24	0.06	3.51	0.17	80.96	0.05	0.16	21.89	0.72	17.61	7.24	0.44	0.01	1.93	368.78	89.16	1.39	1.74	3.34	2.21	0.03	0.22	5.55	19.67	1.06	54.73	7.99	9.66	0.85	0.09	0.08	0.64	0.25	0.39	76.71	3.09	1.55

How to use the data

You can load the dataset using the load_dataset function from the datasets library. When loading the dataset, you can specify which Language you would like to access. Additionally, you will need to use an access token, which you can obtain here, to access the dataset.

from datasets import load_dataset
dataset_subset = load_dataset("ARTPARK-IISc/Vaani-transcription-part", "Angika",token=<access_token>)

To learn more, please click here

Citation

If you use this data, please cite the following:

@misc{vaani2025,
  author       = {Indian Institute of Science (IISc) and ARTPARK},
  title        = {Capturing the Language Landscape for an Inclusive Digital India},
  howpublished = {\url{https://vaani.iisc.ac.in/}},
  year         = {2025}
}

Contact Us

We are always open to new ideas and collaborations. Feel free to reach out to us at [email protected]