Picture for Roman Korostik

Roman Korostik

Generative Speech Foundation Model Pretraining for High-Quality Speech Extraction and Restoration

Add code
Sep 25, 2024
Viaarxiv icon

Schrödinger Bridge for Generative Speech Enhancement

Add code
Jul 22, 2024
Viaarxiv icon

Text-only domain adaptation for end-to-end ASR using integrated text-to-mel-spectrogram generator

Add code
Feb 27, 2023
Viaarxiv icon

You Do Not Need More Data: Improving End-To-End Speech Recognition by Text-To-Speech Data Augmentation

Add code
May 14, 2020
Figure 1 for You Do Not Need More Data: Improving End-To-End Speech Recognition by Text-To-Speech Data Augmentation
Figure 2 for You Do Not Need More Data: Improving End-To-End Speech Recognition by Text-To-Speech Data Augmentation
Figure 3 for You Do Not Need More Data: Improving End-To-End Speech Recognition by Text-To-Speech Data Augmentation
Figure 4 for You Do Not Need More Data: Improving End-To-End Speech Recognition by Text-To-Speech Data Augmentation
Viaarxiv icon