copydata.ai
FounderEngineer

~/papers

The reading database.

Every paper and book I’ve actually read, and what made me read it. No reading lists copied from someone else. 5 entries, published 2010 to 2022.

Topic

5 of 5

YearPaperTopicRead for
2022Robust Speech Recognition via Large-Scale Weak Supervision

Radford, Kim, Xu, Brockman, McLeavey, Sutskever · arXiv 2212.04356

The Whisper paper. Mun's catalog ships Whisper checkpoints, and its translation mode is Whisper's.

Speechmun
2020pyannote.audio: neural building blocks for speaker diarization

Bredin, Yin, Coria, Gelly, Korshunov, Lavechin, et al. · ICASSP 2020 · arXiv 1911.01255

Read for Mun's diarization design: one pinned local pipeline, run as a separate stage, never a hosted service.

Speechmun
2018Competitive Programmer's Handbook

Laaksonen · Book · CSES

Working through it chapter by chapter, with the CSES problem set as homework.

AlgorithmsCompetitive programming
2015Deep learning

LeCun, Bengio, Hinton · Nature 521

The source for a Research Methods exercise on quoting, paraphrasing and citing properly.

Machine learningResearch Methods
2010ZooKeeper: Wait-free Coordination for Internet-scale Systems

Hunt, Konar, Junqueira, Reed · USENIX ATC 2010

Read alongside the ZooKeeper lab, after writing an HTTP client and server by hand in Java.

Distributed systemsDistributed Systems, ZooKeeper lab

Entries live in src/content/papers.json. One object per paper.