I write notes alongside code to capture what I learn while building, while it is still fresh. My writing includes audio representations, cover-song identification, sequence alignment, and honest evaluation of retrieval, connecting to my work on music audio and ai work.
I write notes next to the code.
I write down what I learn while building, while it is still fresh.
- The subjects I write about include audio representations, cover-song identification, sequence alignment, and how to evaluate retrieval honestly.
I have published Audio Explorer: Building the Audio AI Pipeline from First Principles at https://hey-shiv.github.io/blogs/Audio%20Explorer/.
The Audio Explorer post is dated August 2026.
I categorize the Audio Explorer post as audio AI + DSP.
The post traces audio processing from raw samples through spectrograms, Mel features, and MFCCs, and also covers beat tracking.