Govinda Madhava BS

• Co-developed idtap - an open-source Python API that enables computational research, analysis & auto-transcripiton of Music. • Developing a DSP-based auto transcription system to convert music recordings into novel trajectory-based notation, tackling challenges in pitch tracking and melodic segmentation.

1. Develop cross-platform C++ applications and User Interfaces powering Tanpura Droid (1M+ downloads) and Swar Shala, directly impacting over a million users and enabling Indian music learning, composition, and production at scale. 2. Translated research into production by architecting an end-to-end ML pipeline delivering real-time, low-latency pitch shifting (±12 semitones) using Google Magenta DDSP and Granular Audio Synthesis. 3. Applied time-frequency masking reducing audio bleeding in binaural datasets, improving Signal-to-Noise Ratio by 68%. 4. Optimized real-time user pitch tracking by 29% by improving Circular Buffer management and Dynamic Time Warping (DTW).

1. Curated a music dataset comprising 4 hr of high quality audio, metadata, rendition-wise digitized notations. 2. Achieved 68% accuracy developing a Bi-LSTM architecture, the first automated transcription system for Indian classical vocal performances, converting raw audio signals to pitch contours and schematic notation.

- Dean’s Fellowship (2025-26): 1 of 2 students selected for the award from the cohort of 50+ Georgia Tech Music Technology graduate students. - Relevant Courses: Audio Content Analysis, Audio DSP, Integrated Music & Multimedia, Conversational AI - Graduate Student Researcher @Music Informatics Lab, advised by Prof. Alexander Lerch. - Delivered a guest lecture for the Music Dept, Auburn University, AL on Indian Music: Creative Aspects and Practice.