Artificial Intelligence & ML
JSON
Version 2.1.0
Updated Oct 2026
Multi-Lingual Indic Conversational Dialogue Corpus (12 Languages)
High-quality annotated conversational dataset covering 12 major Indian languages for fine-tuning LLMs and chatbots.
Dataset Overview & Methodology
### Overview
Designed specifically for natural language understanding (NLU), instruction fine-tuning, and semantic retrieval in Indian languages including Hindi, Bengali, Tamil, Telugu, Marathi, Gujarati, Kannada, Malayalam, Punjabi, Odia, Assamese, and Indian English.
### Attributes
- 1,200,000 multi-turn conversation pairs
- Sentiment labels, intent classifications, and entity extractions
- Cleaned of toxic text and validated by native speakers.
Designed specifically for natural language understanding (NLU), instruction fine-tuning, and semantic retrieval in Indian languages including Hindi, Bengali, Tamil, Telugu, Marathi, Gujarati, Kannada, Malayalam, Punjabi, Odia, Assamese, and Indian English.
### Attributes
- 1,200,000 multi-turn conversation pairs
- Sentiment labels, intent classifications, and entity extractions
- Cleaned of toxic text and validated by native speakers.
Data Schema & Column Dictionary
5 Attributes| Field Name | Data Type | Description / Valid Values |
|---|---|---|
dialogue_id |
varchar(32) | Unique dialogue identifier |
language |
varchar(20) | Language name and ISO code |
turns |
array[object] | Sequence of role:speaker, text, and intent |
sentiment |
varchar(15) | Positive, Neutral, Negative |
domain |
varchar(30) | Banking, Support, Health, Travel |
Sample Data Preview
Verified SampleThe sample rows below demonstrate the normalized structure of this dataset. The complete paid dataset contains the full volume of records.
| dialogue_id | language | turns | sentiment | domain |
|---|---|---|---|---|
| dlg_hi_00192 | Hindi (hi) | [{"role":"user","text":"\u092e\u0947\u0930\u093e \u092c\u0948\u0902\u0915 \u092c\u0948\u0932\u0947\u0902\u0938 \u0915\u093f\u0924\u0928\u093e \u0939\u0948?"},{"role":"assistant","text":"\u0915\u0943\u092a\u092f\u093e \u0905\u092a\u0928\u093e \u0916\u093e\u0924\u093e \u0928\u0902\u092c\u0930 \u0926\u0930\u094d\u091c \u0915\u0930\u0947\u0902\u0964"}] | Neutral | Banking |
Included Deliverable Files
Full Dataset Bundle (JSON)
Original file: indic-conversational-dialogue-corpus-2.1.0.csv • 2.99 KB • MIME: text/csv
Version Release History
-
v2.1.0 28 Oct 2026
Initial marketplace release with verified schema and data checks.
License & Permitted Usage
AI/ML Commercial Training License: Authorizes direct ingestion into foundational model pre-training, fine-tuning, and embedding generation.
One-Time Purchase Price
₹1,299.00
INR
Payment Type:
One-time payment
Total Records:
1,200,000
File Size:
150.00 MB
Format:
JSON
Cashfree Payments
Official Hosted Checkout • UPI, Cards, NetBanking
Includes lifetime download access to purchased dataset version.
Related Datasets in Artificial Intelligence & ML
Artificial Intelligence & ML
JSON
Test NLP Corpus
0 B
50,000 rows
v1.0.0
Preview