Research group
Gaash Lab
NIT Srinagar
The NIT Srinagar group behind KATHE 2026, a shared task on English→Kashmiri translation, and Bolbosh, an open Kashmiri text-to-speech system with its training data. It also released a Kashmiri fine-tune of Sarvam-Translate and a Matcha-TTS voice.
- 1
- Papers
- 2
- Datasets
- 3
- Models
- 1
- Tools
Focus
Papers 1
Datasets 2
Bolbosh Kashmiri TTS Dataset — Gaash Lab
About 40,000 Kashmiri utterances with transcripts (33,182 train, 4,542 validation, 2,272 test) curated from IndicVoices-R and RASA for text-to-speech; the training data behind Bolbosh. Mostly spontaneous speech; the RASA part keeps its original licence.
KATHE 2026 — English→Kashmiri Translation Challenge Data
Data for KATHE 2026, the national English→Kashmiri machine translation challenge on Kaggle organised by Gaash Lab, NIT Srinagar, with the Bureau of Indian Standards and the University of Kashmir. The test set is 1,730 short everyday English sentences with hidden Kashmiri references, scored by the geometric mean of BLEU and chrF++; 38 teams took part. The finale was held at NIT Srinagar on 21 August 2026.
Models 3
Bolbosh — Kashmiri TTS
Script-aware flow-matching TTS purpose-built for Kashmiri, with a 272-grapheme vocabulary covering the language's diacritics and vowel distinctions. Current state of the art for Kashmiri speech synthesis at MOS 3.63.
Matcha-TTS Kashmiri — Gaash Lab
A Kashmiri text-to-speech model based on Matcha-TTS, from Gaash Lab, NIT Srinagar.
Sarvam Kashmiri (fine-tuned) — Gaash Lab
A Sarvam language model fine-tuned for Kashmiri by Gaash Lab, NIT Srinagar.
Tools 1
Worked with
- Tajamul Ashraf
- Burhaan Rasheed Zargar
- Saeed Abdul Muizz
- Ifrah Mushtaq
- Nazima Mehdi
- Iqra Altaf Gillani
- Aadil Amin Kak
- Janibul Bashir
Every work above is a record in the Kashmiri NLP resource hub, credited as published; profile details come from github.com, arxiv.org, huggingface.co. Something missing or wrong? Submit it.