Śrāvaṇa Māsa Māhātmya
A month of daily recordings on a text that appears in none of the 488,149 verses we index, and of which no digital exemplar has been found. There is no edition here to read along with, and no second reciter to check against. What there is: 959 passages of chant, each one placed in its recording to the second, so that a month of Śrāvaṇa can be heard without being read.
The atlases on this site are text-first: a printed edition is the spine, a recording is hung on it, and the point worth making is that the two were authenticated separately. The Tulā Kāverī page had no edition either, and answered that with a second reciter — two witnesses agreeing verse for verse is evidence about a text produced without the text. Here there is one witness, so neither answer is available.
So this is an anukramaṇikā: an index to a recording rather than to a book. The Devanagari below is one transcription pass's decode of what was chanted — a finding aid, not a reading text, and not an edition of anything. Read it to recognise a verse and press the button; do not cite it.
Both panels are the same probe
(scripts/attribute.py), the same reciter and the same evidence —
each day's transcript asked, on its own, which of the indexed divisions it
most resembles. The strength of the winning match does not separate
the two series and is deliberately not drawn: Kārtika's median best rate is
0.63% of a day's segments and Śrāvaṇa's is 0.28%, which would tell a reader
nothing. What separates them is whether the same answer comes back
tomorrow.
Each heading is one recording, titled by the reciter himself with the tithi and the vrata of that day. Under it, in the order they are chanted, the Sanskrit passages of that day. Press one to hear it.