A Discourse-level Multi-scale Prosodic Model for Fine-grained Emotion Analysis

TL;DR

This paper introduces a discourse-level multi-scale prosodic model for fine-grained emotion analysis, leveraging multi-scale text features and prosodic embeddings to improve expressive speech synthesis and emotional prosody prediction.

Contribution

It proposes a novel D-MPM model that exploits multi-scale discourse-level text to predict prosodic features, enhancing emotional speech synthesis and analysis.

Findings

Multi-scale text improves prosodic feature prediction.

Discourse-level text enhances speech coherence and user experience.

Synthesized speech quality surpasses style transfer in some evaluations.

Abstract

This paper explores predicting suitable prosodic features for fine-grained emotion analysis from the discourse-level text. To obtain fine-grained emotional prosodic features as predictive values for our model, we extract a phoneme-level Local Prosody Embedding sequence (LPEs) and a Global Style Embedding as prosodic speech features from the speech with the help of a style transfer model. We propose a Discourse-level Multi-scale text Prosodic Model (D-MPM) that exploits multi-scale text to predict these two prosodic features. The proposed model can be used to analyze better emotional prosodic features and thus guide the speech synthesis model to synthesize more expressive speech. To quantitatively evaluate the proposed model, we contribute a new and large-scale Discourse-level Chinese Audiobook (DCA) dataset with more than 13,000 utterances annotated sequences to evaluate the proposed model. Experimental results on the DCA dataset show that the multi-scale text information effectively helps to predict prosodic features, and the discourse-level text improves both the overall coherence and the user experience. More interestingly, although we aim at the synthesis effect of the style transfer model, the synthesized speech by the proposed text prosodic analysis model is even better than the style transfer from the original speech in some user evaluation indicators.