“Stop thinking about (art) works as objects and start thinking about them as triggers for experiences. What makes a work (of art) good for you is not something that s already inside it but something that happens inside you.”- Brian Eno
(2024) Challenge - ASVSpoof5
Enhancing SOTA Speech Deepfake Detection with GEMAPS Acoustic Features. Entry to the ASVspoof5 challenge.
(2023) Challenge - Kaggle Speech Recognition
Experimented with timbre perturbation as data augmentation for a Wav2Vec2. Entry to Kaggle’s Bengali.AI ASR challenge.
(2022-2024) Codebase - Cross-Speaker Style Transfer
A fork from Coqui-AI (🐸TTS) used to research about expressive TTS.
(2022) Codebase - TTS Objective Metrics
A compilation of the objective metrics used in several text-to-speech (TTS) papers.
(2022) Hackathon - The Sound of AI
a i t m o s p h e r i c: Generating soundscapes with VQ‑VAEs for compositional use and inspiration. Entry to the 1st Sound of AI Hackathon.