Can We Generate Emotional Pronunciations for Expressive Speech Synthesis?
By Marie Tahon, Gwénolé Lecorvé, Damien Lolive. 2020
18 citations recorded by the source corpus.
In the field of expressive speech synthesis, a lot of work has been conducted on suprasegmental prosodic features while few has been done on pronunciation variants. However, prosody is highly related to the sequence of phonemes to be expressed. This article raises two issues in the generation of emotional pronunciations for TTS systems. The first issue consists in designing an automatic pronunciation generation method from text, while the second issue addresses the very existence of emotional pronunciations through experiments conducted on emotional speech. To do so, an innovative pronunciation adaptation method which automatically adapts canonical phonemes first to those labeled in the corpus used to create a synthetic voice, then to those labeled in an expressive corpus, is presented. This method consists in training conditional random fields pronunciation models with prosodic, linguistic, phonological and articulatory features. The analysis of emotional pronunciations reveals strong dependencies between prosody and phoneme assimilation or elisions. According to perceptual tests, the double adaptation allows to synthesize expressive speech samples of good quality, but emotion-specific pronunciations are too subtle to be perceived by testers.
Subfields: Artificial Intelligence, Experimental and Cognitive Psychology
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Björn Schuller
- Zhaoxi Mu
- Xinyu Yang
- Panagiotis Tzirakis
- Ruibo Fu
- Yizhuo Dong
- Florian Eyben
- Maja Pantic
In Experimental and Cognitive Psychology
- Panagiotis Tzirakis
- Björn Schuller
- Zhaoxi Mu
- Xinyu Yang
- Yizhuo Dong
- Florian Eyben
- Maja Pantic
- Anton Batliner
Citations in the corpus
- An Overview of Affective Speech Synthesis and Conversion in the Deep Learning Era (2022, background): “For example, Tahon et al. [51] investigated the potential of generating emotional pronunciations to improve expressivity.”
- Review of end-to-end speech synthesis technology based on deep learning (2021, methodology): “[203] added a pronunciation adaptive framework based on CRF between text front-end and TTS model to generate different styles of speech.”