Felipe Espic
Senior Applied Research Scientist at Spotify
City of Edinburgh, Scotland, United Kingdom
Summary
🤩
Rockstar🎓
Top SchoolFelipe Espic is a Senior Applied Research Scientist based in Edinburgh with a decade of experience focused on audio and speech research engineering. He combines deep academic training—a PhD from the University of Edinburgh and an MS in Music Engineering Technology—with industry research roles at Spotify and Apple, translating cutting-edge speech models into production-ready systems. His open-source contributions to the well-regarded Merlin speech synthesis project include integrating the MagPhase vocoder and REAPER F0 extraction, work that directly improved voice conversion and generation pipelines. Felipe’s background spans hands-on audio DSP, teaching, and studio engineering, giving him a rare blend of signal-processing rigor and practical audio production sensibility. He excels at bridging research prototypes and scalable implementations, often extending core tooling to enable new functionality such as constant frame rate processing. Colleagues describe him as a problem-solver who leverages both academic depth and product-focused engineering to push speech tech forward.
10 years of coding experience
12 years of employment as a software developer
Doctor of Philosophy - PhD, Doctor of Philosophy - PhD at The University of EdinburghMaster of Science in Music Engineering Technology, Master of Science in Music Engineering Technology at University of MiamiBachelor’s Degree in Engineering Sciences, Bachelor’s Degree in Engineering Sciences at Universidad Tecnológica de Chile INACAP
Spanish, English
Github Skills (10)
machine-learning10
text-to-speech10
speech-synthesis10
deep-learning10
voice-conversion10
tensorflow10
python10
numpy9
keras9
theano8
Programming languages (7)
Github contributions (5)
CSTR-Edinburgh/merlin
Oct 2017 - Apr 2018
This is now the official location of the Merlin project.
Role in this project:
ML Engineer
ML EngineerContributions:49 commits, 9 PRs, 11 pushes in 6 months
Contributions summary:Felipe integrated REAPER for F0 extraction and implemented feature extraction for the MagPhase vocoder, which is crucial for speech synthesis and voice conversion tasks within the Merlin project. They further integrated MagPhase vocoder into the Merlin python code, adding new configuration options and making modifications to the generation functions. The user also adapted the codebase to work with constant frame rate mode, enabling new functionalities for the project. The user's work heavily involves the core functionality of the Merlin project related to speech synthesis and voice conversion using MagPhase vocoder.
pythondeep-learningvoice-conversiontheanospeech-synthesis
felipeespic/matide
Jan 2018 - Jun 2020
Contributions:27 commits, 2 PRs, 23 pushes in 2 years 5 months
signalpythonimage-signal-processingimage-processingide
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.