T.V. Raman's AsTeR (1994) rendered mathematical structure as pitched, spatial audio — raising the voice for superscripts and receding nested subexpressions into audio 'distance'

AsTeR — Audio System for Technical Readings, the subject of T. V. Raman's 1994 Cornell PhD thesis — did not read mathematics as a flat linear string. It rendered the structure of an expression into non-speech audio dimensions layered on top of the spoken symbols. Two mappings are documented in the thesis:

• Pitch encodes verticality. A superscript raises the reading voice: "a change along the audio dimension shown in fig:superscript produces a higher pitched voice." The two-dimensional spatial fact of an exponent sitting above the baseline becomes an audible fact.
• Nesting recedes into audio "distance." As subexpressions nest deeper, the voice characteristic shifts so the reader hears the inner term "falling off into the distance." A reduced step along the average-pitch and "head-size" dimensions further disambiguates, for example, a subscript inside a superscript.

The design treats an equation the way a rendering engine treats a document: it has audio formatting — an audio analog of the visual cues (size, position, style) that a sighted reader parses pre-attentively. This makes AsTeR the direct mechanical ancestor of later position-to-pitch schemes in math sonification; the same "pitch = vertical position" mapping reappears in claim-stereomath-2024-maps-equation-position-to-stereo-pitch-and-earcons three decades later, a convergence tracked at observation-math-sonification-pitch-for-vertical-recurred-1994-2024.

AsTeR is one modality-answer in the larger pattern of blind mathematicians building their own means to work — speech for Pontryagin, touch for Saunderson, and here, structured sound (see moc-blind-mathematicians-working-methods). Raman, blind since fourteen, built it himself — an instance of observation-blind-math-accessibility-built-by-users-not-toolmakers.