Computational Social Linguistics for Telugu Cultural Preservation: Novel Algorithms for Chandassu Metrical Pattern Recognition

Boddu Sri Pavan; Boddu Swathi Sree

cs.CLSep 23, 2025

Computational Social Linguistics for Telugu Cultural Preservation: Novel Algorithms for Chandassu Metrical Pattern Recognition

Authors: Boddu Sri Pavan, Boddu Swathi Sree

Organizations: Independent Researcher · Department of Computer Science and Engineering RGUKT, Nuzvid

Abstract

This research presents a computational social science approach to preserving Telugu Chandassu, the metrical poetry tradition representing centuries of collective cultural intelligence. We develop the first comprehensive digital framework for analyzing Telugu prosodic patterns, bridging traditional community knowledge with modern computational methods. Our social computing approach involves collaborative dataset creation of 4,651 annotated padyams, expert-validated linguistic patterns, and culturally-informed algorithmic design. The framework includes AksharamTokenizer for prosody-aware tokenization, LaghuvuGuruvu Generator for classifying light and heavy syllables, and PadyaBhedam Checker for automated pattern recognition. Our algorithm achieves 91.73% accuracy on the proposed Chandassu Score, with evaluation metrics reflecting traditional literary standards. This work demonstrates how computational social science can preserve endangered cultural knowledge systems while enabling new forms of collective intelligence around literary heritage. The methodology offers insights for community-centered approaches to cultural preservation, supporting broader initiatives in digital humanities and socially-aware computing systems.

Figures & tables

Aksharam Token Count	Ganam Name	LaghuvuGuruvu Pattern
1	LA	\|
1	GA	U
2	LAA	\|\|
2	VA	\|U
2	HA	U\|
2	GAA	UU

Table 1: Ganam classification by aksharam token count and laghuvu-guruvu pattern

Prosodic Class	Padyam Types
Vruttamu	Vutpalamaala, Champakamaala, Saardulamu, Mattebhamu
Jaathi	Kandamu
Vupajaathi	Teytageethi, Aataveladi, Seesamu

Table 2: Telugu prosodic classes and types in Chandassu dataset

Figure 1: Distribution of padyams across satakams in Chandassu dataset

Prosodic Class	Padyam Count
Vruttamu	1625
Jaathi	683
Vupajaathi	2343

Table 3: Distribution of padyams by prosodic class in Chandassu dataset

Padyam Type	Padyam Count
Vutpalamaala	329
Champakamaala	389
Saardulamu	290
Mattebhamu	617
Kandamu	683
Teytageethi	676

Table 4: Distribution of padyams by type in Chandassu dataset

Figure 2: Telugu character set used by the proposed framework

Figure 3: Proposed computational framework architecture for Telugu metrical poetry analysis

Figure 4: Performance evaluation across prosodic classes

Figure 5: Performance evaluation across padyam types

Prosodic Class	$c_{na}$	$c_{np}$	$c_{gk}$	$c_{yt}$	$c_{pr}$	$C$
Vruttamu	99.43	99.94	93.78	81.38	96.12	94.13
Jaathi	-	100.00	92.85	73.94	90.78	89.39
Vupajaathi	-	99.91	94.12	78.21	-	90.75

Table 5: Performance evaluation by prosodic class (values in %)

Padyam Type	$c_{na}$	$c_{np}$	$c_{gk}$	$c_{yt}$	$c_{pr}$	$C$
Vutpalamaala	98.97	99.70	91.87	77.36	94.60	92.50
Champakamaala	99.54	100.00	96.02	83.48	95.69	94.95
Saardulamu	99.35	100.00	94.88	79.66	97.59	94.29
Mattebhamu	99.63	100.00	92.89	83.02	96.52	94.41
Kandamu	-	100.00	92.85	73.94	90.78	89.39
Teytageethi	-	99.85	92.85	76.81	-	89.84

Table 6: Performance evaluation by padyam type (values in %)

Cultural Heritage

Cultural heritage uses AI to document, analyze, preserve, and interpret artifacts, sites, archives, and living traditions.

See all 57 papers

Oct 17, 2026

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

Sanskrit verse recitation is governed by

\textit{Chandas}

, a strict prosodic framework defined by rhythmic

\textit{Laghu}

(light) and

\textit{Guru}

(heavy) syllables. Because existing Sanskrit speech datasets prioritize ASR or conversational TTS, they lack the metrical fidelity required for authentic verse synthesis. We introduce

\textbf{

ShlokaDhara

}

, a 43-hour, multi-speaker Sanskrit speech corpus capturing diverse meters, recitation styles, and melodic variations. We specifically focus on

\textit{Text-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

\textit{Laghu-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/

Harshal Kausadikar, Onkar Kishor Susladkar, Prathmesh Santosh Choudhari +3

Independent Researcher · University of Illinois at Urbana-Champaign · University of Florida +2

Speech Processing TTS Synthesis Speech Corpus Construction TTS Evaluation

Ask PDF

Mar 13, 2026

Perpetual Dialogues: A Computational Analysis of Voice-Guitar Interaction in Carlos Paredes's Discography

Computational musicology enables systematic analysis of performative and structural traits in recorded music, yet existing approaches remain largely tailored to notated, score-based repertoires. This study advances a methodology for analyzing voice-guitar interaction in Carlos Paredes's vocal collaborations - an oral-tradition context where compositional and performative layers co-emerge. Using source-separated stems, physics-informed harmonic modelling, and beat-level audio descriptors, we examine melodic, harmonic, and rhythmic relationships across eight recordings with four singers. Our commonality-diversity framework, combining multi-scale correlation analysis with residual-based detection of structural deviations, reveals that expressive coordination is predominantly piece-specific rather than corpus-wide. Diversity events systematically align with formal boundaries and textural shifts, demonstrating that the proposed approach can identify musically salient reorganizations with minimal human annotation. The framework further offers a generalizable computational strategy for repertoires without notated blueprints, extending Music Performance Analysis into oral-tradition and improvisation-inflected practices.

Gilberto Bernardes, Nádia Moura, António Sá Pinto

INESC TEC, Faculty of Engineering, University of Porto · CEIS20, Faculty of Arts and Humanities, University of Coimbra

Computational Musicology Music Information Retrieval

Ask PDF

Mar 25, 2026

Pingala: Prosody-Aware Decoding for Sanskrit Poetry Generation

Poetry generation in Sanskrit typically requires the verse to be semantically coherent and adhere to strict prosodic rules. In Sanskrit prosody, every line of a verse is typically a fixed length sequence of syllables adhering to prescribed binary patterns of syllable weights. We observe that instead of treating a verse as a monolithic sequence, segmenting them as grouped-lines leads to significant improvement in semantic coherence by 10% with comparable metrical adherence. Specifically, Pingala, our proposed decoding approach is designed to encourage every line to have well-formed words and our token selection biases the model towards it by preferring longer tokens. Writing in Sanskrit follows phonemic orthography, hence using a phonetically aware transliteration scheme, SLP1, increased the metrical alignment by 46% with comparable semantic similarity, for a instruction fine-tuned large language models like Phi-4. We also introduce a new approach for reference-free evaluation using cross-encoders which achieved better alignment with true poetry instances.

Manoj Balaji Jagadeeshan, Atul Singh, Nallani Chakravartula Sahith +2

Indian Institute of Technology, Kharagpur · Oriﬂow

Text Generation Language Model Decoding Controllable Text Generation Grammar-Constrained Decoding Language Model Generation Evaluation

Ask PDF

Abstract

Cultural Heritage

Explore similar work

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

Perpetual Dialogues: A Computational Analysis of Voice-Guitar Interaction in Carlos Paredes's Discography

Pingala: Prosody-Aware Decoding for Sanskrit Poetry Generation