Use the catalogue without reading a page.

Everything on this site is available to software in the form it prefers. All of it is read-only and needs no key.

When to use it

You need two-speaker conversational speech in Hindi, Tamil or Marathi today; other Indian languages are collected to a brief.
You are building or evaluating a voice agent and want calls between a person and an AI voice agent, each on its own channel at 48 kHz, in Indian languages: play the sample release (status=preview) and commission the collection in your languages and scenarios.
You are training or evaluating a full-duplex speech-to-speech model (Moshi, PersonaPlex and similar) and need dual-channel audio, where each speaker is on a separate channel. Filter with channels=dual.
You need speech recognition data with natural code-switching: transcripts are in native script with English words kept as spoken, and each dataset reports the share of Latin-script words across the whole set.
You need call-centre conversations (insurance, consumer surveys, telecom, delivery, e-commerce, banking) for a voice agent, or utterance clips, speaker-turn files and two-channel calls built from them.
You want to judge fitness before contacting anyone: every dataset has audio figures and a conversation profile measured across the whole set, and a playable sample.

When not to

You need a free or openly licensed dataset. Everything here is licensed and quoted per use.
You need studio-quality speech for a text-to-speech voice. This is 16 kHz conversational audio.
You need read speech, single-speaker audio, or wake words. The catalogue is two-party conversation and what is built from it.
You need images, documents or preference data today. Those are collections Lokah scopes on request.

Interfaces

llms.txtThe Kenpath Labs site in one file, with a Lokah section listing every dataset. /llms.txt
llms-full.txtThe same, with every dataset's full record and the formats guide. /llms-full.txt
JSON APIList and filter datasets. One record at /api/lokah/datasets/{slug}. /api/lokah/datasets
OpenAPI 3.1Machine-readable description of the JSON API. /lokah/openapi.json
MCP serverModel Context Protocol over streamable HTTP: search_datasets, get_dataset, get_sample, list_languages, and send_enquiry to reach us. /api/lokah/mcp
Enquiry endpointPOST a JSON enquiry on a person's behalf: a quote, a sample bundle, a collection to a brief, or a question. A person replies by email. /api/lokah/enquiry
CroissantMLCommons Croissant 1.0 metadata with the responsible-AI block, per dataset. /lokah/datasets/{slug}/croissant.json
schema.orgA Dataset JSON-LD block in every dataset page, and a DataCatalog on /datasets. /lokah/datasets/{slug}
MarkdownAny page as Markdown: send Accept: text/markdown, or add .md. Dataset pages open with a Hugging Face dataset card header. /lokah/datasets/{slug}.md
SitemapEvery page of the site. /sitemap.xml

How to use it

List datasets: GET https://kenpathlabs.com/api/lokah/datasets with any of language, channels (dual or mono), collection (call-centre or general-conversation), min_hours, q. Each record carries kind (conversations, asr, diarization or duplex), hours, audioQuality and fitFor.
Read one: GET https://kenpathlabs.com/api/lokah/datasets/{slug}. measured holds the whole-set figures (files, hours, speakers, snrMedianDb, bandwidth, profile); sample holds what was measured from the sample conversations; gaps lists what is not held.
Check fitness for full-duplex training: channels must be "dual", then read measured.profile.twoChannel (IPUs, pauses, gaps, overlaps and backchannels per minute, the median floor-transfer offset, channel isolation), measured across every call in the set.
Judge audio quality from measured.snrMedianDb across the set (under 18 dB noisy, over 30 clean) and measured.bandwidth. The sample adds sample.quality.dnsmos (P.835 background, speech and overall, 1 to 5).
Hear it: sample.specimens lists the public excerpts of a conversation set; utterances.clips lists them for a speech-recognition set. Their URLs and transcripts are in the Croissant file's distribution.
Get a sample or a price: send the person to https://kenpathlabs.com/lokah/datasets/{slug}#sample. A sample bundle arrives by email under the Evaluation Licence; a licence for the full set is quoted per use. There is no public price and no checkout.
Reach us yourself: once the person agrees, POST https://kenpathlabs.com/api/lokah/enquiry with name, email, intent (quote, sample, collection or question), and dataset or message as needed, or call the MCP tool send_enquiry. The answer says what happens next; a person replies to the email given. The same address for anything else: hello@kenpathlabs.com.

Find dual-channel Hindi data

curl "https://kenpathlabs.com/api/lokah/datasets?language=hi-IN&channels=dual"

MCP client configuration

{ "mcpServers": { "lokah": { "type": "http", "url": "https://kenpathlabs.com/api/lokah/mcp" } } }

Any page as Markdown

curl -H "Accept: text/markdown" https://kenpathlabs.com/lokah/datasets/hindi-call-centre-insurance

Send an enquiry on a person's behalf

curl -X POST https://kenpathlabs.com/api/lokah/enquiry \
  -H "Content-Type: application/json" \
  -d '{"name":"Asha Rao","email":"asha@example.com","intent":"quote","dataset":"hindi-call-centre-insurance","message":"200 hours for a full-duplex model","agent":"my-assistant"}'

Reaching us

Agents are welcome to contact us directly on a person's behalf, once that person has agreed.

POST https://kenpathlabs.com/api/lokah/enquiry with JSON: name, email (where we reply), intent (quote, sample, collection or question), and dataset (slug) or message as needed. Answer: 202 with what happens next. Documented in https://kenpathlabs.com/lokah/openapi.json.

Or the MCP tool send_enquiry on https://kenpathlabs.com/api/lokah/mcp, same fields.

Or email hello@kenpathlabs.com. A person replies. Nothing is priced, licensed or signed automatically; every licence is quoted per use and agreed by people.