bolAIndia

A growing atlas of India’s spoken voices

India does not speak in a single voice. It speaks through a vast and vibrant mosaic of languages, regional accents, dialects, and everyday expressions—across cities, towns, and rural communities.

bolAIndia is a large-scale, expanding collection of Indian speech built to help speech AI hear that richness. It brings together conversational, read, educational, and broadcast audio, including code-switching and regional varieties, in a consistent, filterable format.

Our ambition is to help researchers build speech technologies that serve more of India: across linguistic, geographic, and social boundaries. Breadth is a continuing goal, not a claim of complete or demographically balanced coverage.

The collection

Total audio hours	Transcribed hours	Untranscribed hours
190.06K	145.07K	44.98K

K = thousand · M = million. Display values are rounded independently; detailed figures are in the data guide.

Hours count published dataset rows, excluding separately preserved source archives. Transcribed includes human and automated annotations, including labelled unintelligible examples; it does not imply human verification.

India, in sound

A regional view of the languages represented in the collection—not a map of recording locations. Seven distinct shades combine published language hours with historical Census mother-tongue profiles. Fine lines show district boundaries; colour remains state-level, not district recording coverage. English remains in the language charts rather than being assigned a regional footprint.

Frozen view: refreshed only on request. The figures and bar charts continue to update independently. Map methodology, geographic limitations and snapshot details.

Languages of India, in sound

Primary = up to 15 largest labelled languages by published audio hours; secondary = all remaining languages. These are display groups, not a ranking of linguistic importance. Groups and values refresh with dataset publication.

Primary languages

Largest collections · published audio hours

Audio hours · linear scale: 0–90.00K

LANGUAGE		HOURS
Hindi		89.81K
Bengali		12.13K
English		9.98K
Tamil		9.15K
Telugu		8.97K
Punjabi		8.83K
Marathi		6.12K
Kannada		5.89K
Urdu		4.11K
Malayalam		4.08K
Gujarati		2.54K
Odia		1.99K
Assamese		1.71K
Nepali		1.49K
Maithili		1.14K

Secondary languages

Smaller collections, shown on separate zero-based linear scales so their bars remain readable. Compare bar lengths within a panel, not across panels.

100–1.00K hours · 20 languages

Audio hours · linear scale: 0–1.00K

LANGUAGE		HOURS
Bhojpuri		793.54
Manipuri		781
Chhattisgarhi		628.4
Sanskrit		600.77
Bodo		598.84
Dogri		559.03
Sindhi		493.3
Chakma		480.11
Santali		474.98
Garo		464.85
Kashmiri		444.21
Konkani		416.24
Nagamese		298.94
Magahi		232.25
Rajasthani		205.33
Mizo		199.19
Marwari		157.01
Garhwali		154.04
Wancho		118.85
Angika		108.87

10–100 hours · 30 languages

Audio hours · linear scale: 0–100

LANGUAGE		HOURS
Karbi		83.7
Bajjika		82.91
Haryanvi		63.22
Halbi		52.72
Kokborok		46.54
Tulu		46.27
Sambalpuri		34.41
Khortha		31.4
Kumaoni		26.99
Sadri		26.51
Surjapuri		24.69
Bilaspuri		22.67
Lotha		21.37
Kullu Pahari		21.22
Mandeali		20.2
Gaddi		20.1
Outer Seraji		19.59
Kangri		19.21
Bhadrawahi		18.18
Mahasu Pahari		17.73
Surgujia		16
Nimadi (x-nimadi)		15.06
Malvani		14.78
Sumi		11.53
Korwa		11.42
Konkani (individual language)		11.26
Bundeli		11.21
Kurukh		10.68
Bhili		10.52
Nimadi (noe)		10.01

1–10 hours · 44 languages

Audio hours · linear scale: 0–10

LANGUAGE		HOURS
Jaunsari		9.51
Idu Mishmi		9.25
Kui		9.14
Angami		8.97
Kota		8.67
Gondi		8.36
Awadhi		7.85
Kinnauri		7.55
Goan Konkani		6.78
Lepcha		6.54
Desia		6.46
Beary		5.93
Kathoriya Tharu		5.74
Chakhesang		4.93
Braj		4.85
Khandeshi		4.73
Western Punjabi		4.55
Ao		4.34
Doteli		4.29
Nyishi		3.92
Rana Tharu		3.56
Rengma		3.48
Chitwania Tharu		3.38
Sikkimese (sip)		3.37
Rongmei		3.01
Sindhi		3.01
Deccan		2.8
Kochila Tharu		2.3
Kanauji		2.14
Tagin		1.97
Bagheli		1.91
Koya		1.91
Bhatri		1.65
Sikkimese (x-sikkimese)		1.48
Malvi		1.37
Powari		1.36
Dhundari (Jaipuri)		1.32
Dorli		1.32
Sangtam		1.31
Kurmali		1.3
Pahadi		1.25
Yimchunger		1.14
Tangkhul		1.08
Shekhawati		1

0.1–1 hours · 29 languages

Audio hours · linear scale: 0–1

LANGUAGE		HOURS
Thethi		0.83
Maria		0.82
Hadothi		0.74
Bagri		0.69
Mewari (mtr)		0.69
Od		0.64
Lambani		0.64
Liangmai		0.56
Sylheti		0.54
Zeme		0.53
Duruwa		0.5
Wagdi		0.5
Galo		0.46
Northern Hindko		0.4
Thadou		0.39
Dogri		0.38
Hajong		0.36
Pahari-Potwari		0.35
Harauti		0.32
Mewati		0.31
Wadiyara Koli		0.29
Mewari (x-mewari)		0.26
Kachi Koli		0.23
Vaiphei		0.23
Phom		0.22
Limbu		0.22
Rajbanshi		0.22
Sirmauri		0.18
Agariya		0.16

0.01–0.1 hours · 4 languages

Audio hours · linear scale: 0–0.1

LANGUAGE		HOURS
Baghati		0.08
Paniya		0.07
Kuki		0.05
Mara		0.05

Language not yet labelled: 12.89K hours, counted in the collection total but not treated as a language above.

K = thousand · M = million. Nonzero bars smaller than one pixel retain a visible mark. Full numerical breakdowns are in the data guide.

Format, filtering, source credits and usage conditions.