Su-śrotā is a Sanskrit ASR system tuned for recitational and śāstric speech, paired with Vāgbodhinī, an interactive chant practice tool, and documented with transparent experiments including failures and a data flywheel that improved real-world performance.
48
No data
5
1
No data
2026-07-19
It provides a rare combination of scholar-grade ASR for Sanskrit, honest reporting of what did and did not work, and a live data flywheel that reduced in-the-wild CER from 24% to 15.6% using consented user data, all while being open-source and deployable.
Sanskrit speech recognition tuned for śāstric / recitational Sanskrit, plus Vāgbodhinī, an interactive chant-practice tool built on it. This repository documents the full set of experiments — including the ones that did not work, which were as instructive as the ones that did.
Headline result (§5b): the model was only saturated on clean studio audio. On real, in-the-wild user audio it had huge headroom (24% CER) — and a small batch of consented user recordings closed a third of it (24% → 16% CER, 86% → 34% WER) with no loss on the studio benchmarks. The data flywheel works.
Off-the-shelf Sanskrit ASR is trained on conversational IndicVoices-style data and degrades badly on recitation and śāstra — dense compounds, sandhi, retroflex/aspirate contrasts, pitch, and long metrical utterances. The goal here was a model and tooling good enough for scholars: accurate on chant and prose, and useful as a practice aid rather than a transcription toy.
Two things shaped everything:
Scholar-grade Sanskrit ASR (finetuned IndicConformer-CTC) + Vāgbodhinī metre-aware chant-practice tool
Similar projects matched by category, topics, and programming language.
Harness Engineering is a methodology for improving coding agent outputs by carefully crafting the environment around them—providing curated context, tools, and executable constraints that encode an organization’s nonfunctional requirements and cumulative lessons.
A 28.9 million parameter language model runs on an $8 ESP32-S3 microcontroller entirely on-device, generating simple stories at about 9.5 tokens per second.
mimic captures traffic from any iOS or web app and automatically generates a Python client library that lets you call the app's API like a regular library.