ZackHodari
@ZackHodariGenerative AI, Large Speech Models, Prosody and Expressivity in Synthetic Voices
Language Breakdown
Lines of code distribution across 8 owned repositories
I-Shaped Developer
I-shapedSpecialist — deep expertise in HTML
Collaboration Network
Global Impact visualization
Repos
21
PRs
0
Growth
+18%
Top Collaborators
No collaborator data yet.
Coding Streak
Contribution activity over the past year
Top Repositories
Data processing tools for preparing speech and labels for training TTS voices
Code for paper titled "Using generative modelling to produce varied intonation for speech synthesis" submitted to the Speech Synthesis Workshop
Code for paper titled "Perception of prosodic variation for speech synthesis using an unsupervised discrete representation of F0" submitted to Speech Prosody
Toolkit for defining and training Text-to-Speech voices in PyTorch
Tools for computing metrics between speech samples
Code relating to paper submitted for review at INTERSPEECH 2018
*BeaqleJS* provides a framework to create browser based listening tests and is purely based on open web standards like HTML5 and Javascript.
Modular NN interface for Tensorflow
WhisperX: Automatic Speech Recognition with Word-level Timestamps (& Diarization)
Sphinx extension that automatically document argparse commands and options
Open Source Impact
Contributions to external projects