Whispered and Lombard Neural Speech Synthesis

TL;DR

This paper explores methods for generating multiple speaking styles, including Lombard and whisper, using limited data, and demonstrates high-quality synthesis and improved intelligibility, leveraging pre-training, signal processing, and style embedding techniques.

Contribution

It introduces a comprehensive comparison of approaches for multi-style speech synthesis, including a novel use of a speaker verification model as a style encoder.

Findings

High-quality speech generation via pre-training and fine-tuning.

Lombard and whisper styles can be effectively synthesized with limited data.

Lombard speech improves intelligibility significantly.

Abstract

It is desirable for a text-to-speech system to take into account the environment where synthetic speech is presented, and provide appropriate context-dependent output to the user. In this paper, we present and compare various approaches for generating different speaking styles, namely, normal, Lombard, and whisper speech, using only limited data. The following systems are proposed and assessed: 1) Pre-training and fine-tuning a model for each style. 2) Lombard and whisper speech conversion through a signal processing based approach. 3) Multi-style generation using a single model based on a speaker verification model. Our mean opinion score and AB preference listening tests show that 1) we can generate high quality speech through the pre-training/fine-tuning approach for all speaking styles. 2) Although our speaker verification (SV) model is not explicitly trained to discriminate different speaking styles, and no Lombard and whisper voice is used for pre-training this system, the SV model can be used as a style encoder for generating different style embeddings as input for the Tacotron system. We also show that the resulting synthetic Lombard speech has a significant positive impact on intelligibility gain.