# ShlokaDhara: A Multi-Speaker Sanskrit Speech Corpus for Learning Meter-Aware Prosody in Text-to-Recitation

arXiv: https://arxiv.org/abs/2610.26929 · HTML: https://mlsanity.com/paper/2610.26929/

## Key facts

- No citing papers in the map yet (data through 19 Oct 2026).
- Placed in Vocoders and speech synthesis, one of 11 directions in Speech and audio language models.
- Builds on 15 of its 37 references that are in the map.
- First posted on arXiv 17 Oct 2026.

## Placed in

- Speech and audio language models / [Vocoders and speech synthesis](https://mlsanity.com/speech/vocoders-and-speech-synthesis/)

## Builds on (most cited first, 15 of 15)

- [Conditional Variational Autoencoder with Adversarial Learning for End-to-End Text-to-Speech](https://mlsanity.com/paper/2106.06103/) (2021)
- [UTMOS: UTokyo-SaruLab System for VoiceMOS Challenge 2022](https://mlsanity.com/paper/2204.02152/) (2022)
- [NaturalSpeech 2: Latent Diffusion Models are Natural and Zero-Shot Speech and Singing Synthesizers](https://mlsanity.com/paper/2304.09116/) (2023)
- [XTTS: a Massively Multilingual Zero-Shot Text-to-Speech Model](https://mlsanity.com/paper/2406.04904/) (2024)
- [Natural language guidance of high-fidelity text-to-speech with synthetic annotations](https://mlsanity.com/paper/2402.01912/) (2024)
- [IndicSUPERB: A Speech Processing Universal Performance Benchmark for Indian languages](https://mlsanity.com/paper/2208.11761/) (2022)
- [CLaM-TTS: Improving Neural Codec Language Model for Zero-Shot Text-to-Speech](https://mlsanity.com/paper/2404.02781/) (2024)
- [IndicVoices-R: Unlocking a Massive Multilingual Multi-speaker Speech Corpus for Scaling Indian TTS](https://mlsanity.com/paper/2409.05356/) (2024)
- [Automatic Speech Recognition in Sanskrit: A New Speech Corpus and Modelling Insights](https://mlsanity.com/paper/2106.05852/) (2021)
- [RASMALAI: Resources for Adaptive Speech Modeling in Indian Languages with Accents and Intonations](https://mlsanity.com/paper/2505.18609/) (2025)
- [Chandomitra: Towards Generating Structured Sanskrit Poetry from Natural Language Inputs](https://mlsanity.com/paper/2506.00815/) (2025)
- [Chandojnanam: A Sanskrit Meter Identification and Utilization System](https://mlsanity.com/paper/2209.14924/) (2022)
- [Low-Resource End-to-end Sanskrit TTS using Tacotron2, WaveGlow and Transfer Learning](https://mlsanity.com/paper/2212.03558/) (2022)
- [Bahasa Harmony: A Comprehensive Dataset for Bahasa Text-to-Speech Synthesis with Discrete Codec Modeling of EnGen-TTS](https://mlsanity.com/paper/2410.06608/) (2024)
- [Automatic Speech Recognition for Sanskrit with Transfer Learning](https://mlsanity.com/paper/2501.10024/) (2025)

## Abstract (arXiv)

Sanskrit verse recitation is governed by textitChandas, a strict prosodic framework defined by rhythmic textitLaghu (light) and textitGuru (heavy) syllables. Because existing Sanskrit speech datasets prioritize ASR or conversational TTS, they lack the metrical fidelity required for authentic verse synthesis. We introduce textbftextitShlokaDhara, a 43-hour, multi-speaker Sanskrit speech corpus capturing diverse meters, recitation styles, and melodic variations. We specifically focus on textitText-to-Recite, a synthesis task explicitly requiring both prosodic accuracy and metrical correctness. Benchmarking state-of-the-art TTS architectures via an automated UTMOS (neural MOS) predictor demonstrates that XTTS-v2 achieves the highest perceptual quality. We further validate the XTTS-v2 generated samples using two paradigms: subjective assessments by 16 Sanskrit experts and novel empirical metrics, specifically, syllable-level textitLaghu-Guru duration analysis to objectively quantify structural metrical awareness. Ultimately, this work establishes a robust dataset and evaluation benchmark for prosody-aware computational Sanskrit recitation. Generated Samples: https://shlokadhara.vercel.app/