Exploring Transfer Learning for Low Resource Emotional TTS

Noé Tits, Kevin El Haddad, Thierry Dutoit
1/14/2019
cs.SDcs.CLeess.AS

Official code: Emotional-Text-to-Speech/dl-for-emo-tts (457 stars, MIT, last commit 2024-06-26).

Abstract

During the last few years, spoken language technologies have known a big improvement thanks to Deep Learning. However Deep Learning-based algorithms require amounts of data that are often difficult and costly to gather. Particularly, modeling the variability in speech of different speakers, different styles or different emotions with few data remains challenging. In this paper, we investigate how to leverage fine-tuning on a pre-trained Deep Learning-based TTS model to synthesize speech with a small dataset of another speaker. Then we investigate the possibility to adapt this model to have emotional TTS by fine-tuning the neutral TTS model with a small emotional dataset.

View on arXivView PDF

Code Implementations(1)

Emotional-Text-to-Speech/dl-for-emo-ttsOfficial100%
MIT

:computer: :robot: A summary on our attempts at using Deep Learning approaches for Emotional Text to Speech :speaker:

45742Jupyter NotebookMay 1, 20202 years agoMIT
affective-computingdc-ttsdeep-learningemotional-ttslj-speech+5 more

Similar Papers

Learning Emotional Representations from Imbalanced Speech Data for Speech Emotion Recognition and Emotional Text-to-Speech

84% match2023

Time out of Mind: Generating Rate of Speech conditioned on emotion and speaker

84% match2023

Fine-grained Emotional Control of Text-To-Speech: Learning To Rank Inter- And Intra-Class Emotion Intensities

83% match2023

Plug-and-Play Emotion Graphs for Compositional Prompting in Zero-Shot Speech Emotion Recognition

83% match2025

AffectEcho: Speaker Independent and Language-Agnostic Emotion and Affect Transfer for Speech Synthesis

83% match2023

Attention-based Interactive Disentangling Network for Instance-level Emotional Voice Conversion

82% match2023

Cite this paper

@article{tits2019exploring,
  title  = {Exploring Transfer Learning for Low Resource Emotional TTS},
  author = {Noé Tits and Kevin El Haddad and Thierry Dutoit},
  year   = {2019},
  eprint = {1901.04276},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/1901.04276}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/1901.04276.svg)](https://codeofpaper.com/paper/1901.04276)

Discussion