Companies describe the data they need, where they need it and what they can spend. Posting is free. Verified providers answer with proposals that include a price, and the buyer compares them side by side.
We need 1,500–3,000 hours of spontaneous conversational speech in Hindi and Tamil, 8 kHz telephony and 16 kHz wideband, with verbatim transcripts, speaker metadata (age band, gender, region) and documented speaker consent that covers commercial model training. Balanced across regions and noise conditions. Licence must allow training and fine-tuning of commercial ASR models.
Countries
India🇮🇳India
Delivery
One-time delivery
Budget
₹3M – ₹6M
Posted 3 days ago
3 proposals
How the request pool works
Buyers post for free
Describe the data, countries, delivery and budget. Each request is reviewed before it goes live, usually within one business day.
Providers answer with a price
Verified providers whose data matches send a proposal with an indicative price and delivery details. No price, no proposal.
You choose who to talk to
Buyer details stay private until you shortlist or message a provider. Contracts and payment happen directly between you.