Vāgdhenu — Sanskrit Chant Corpus A single speaker Sanskrit chant (pārāyaṇa) recording corpus — classical ślokas chanted with tradition faithful prosody and metrically aware durations. Training data behind the Vāgdhenu Sanskrit Chant TTS. ~1,467 clips · ~5.3 hours · 24 kHz mono. One reciter (the author); classical śāstra chant, no Vedic svaras. Two configs (two cutting/normalization passes over the recording sessions, largely different verses ): config clips hours sources (sessions) metadata style b 703 2.64 Sumadhvavijaya, MBTN adh 4–7, Vāyustuti, Narasiṃha, Rudra, maṅgala deva · slp1 · kannada · meter · n syll · session · take style a 764 2.70 MBTN adh 1–3 & 7, Sumadhvavijaya, Narasiṃha, Śiva, Vāyustuti, maṅgala deva · slp1 · kannada · duration · session Format (AudioFolder, per config) metadata.csv columns: file name , text devanagari , text slp1 , text kannada (Kannada routed model input), meter (style b), duration , n syll (style b), session , take . Recording protocol (the hard won rules — part of the contribution) One room, one large diaphragm condenser, fixed ~15–20 cm + pop filter, noise floor ≤ −60 dB, 24 bit/48 kHz (→ 24 kHz), peaks ~ −6 dB, lossless WAV. A tanpura/śruti…
Runs entirely in your browser via DuckDB-Wasm — this dataset's real data file is loaded once, then queried locally. Nothing is sent to a server.
We use cookies for essential functionality and analytics. You can accept or reject analytics cookies.Cookie policy