Effects of Large Multi-Speaker Models on the Quality of Neural Speech Synthesis

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F49777513%3A23520%2F24%3A43974114" target="_blank" >RIV/49777513:23520/24:43974114 - isvavai.cz</a>

Výsledek na webu

<a href="https://svk.fav.zcu.cz/download/proceedings_svk_2024.pdf" target="_blank" >https://svk.fav.zcu.cz/download/proceedings_svk_2024.pdf</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Effects of Large Multi-Speaker Models on the Quality of Neural Speech Synthesis

Popis výsledku v původním jazyce

These days, speech synthesis is usually performed by neural models (Tan et al., 2021).A neural speech synthesizer is dependent on a large number of parameters, whose values mustbe acquired during the process of model training. In many situations, the result of trainingcan be improved by fine-tuning a pre-trained model, i.e. using the parameter values of a modelwhich has been trained using different training data to initialize the parameters of the targetmodel before the training process begins (Zhang et al., 2023).In the field of speech synthesis, a pre-trained model is a speech synthesizer which hasbeen trained to synthesize the voice of another speaker. Furthermore, we can use a multi-speakerpre-trained model, which has been trained using speech recordings of multiple speakers, so itshould contain general knowledge about human speech.This paper describes how the number of speakers used to train a pre-trained model affectsthe quality of the final synthetic speech. We used a single-speaker model as well as two multispeakermodels for fine-tuning and we compared the obtained models in a listening test.

Název v anglickém jazyce

Effects of Large Multi-Speaker Models on the Quality of Neural Speech Synthesis

Popis výsledku anglicky

These days, speech synthesis is usually performed by neural models (Tan et al., 2021).A neural speech synthesizer is dependent on a large number of parameters, whose values mustbe acquired during the process of model training. In many situations, the result of trainingcan be improved by fine-tuning a pre-trained model, i.e. using the parameter values of a modelwhich has been trained using different training data to initialize the parameters of the targetmodel before the training process begins (Zhang et al., 2023).In the field of speech synthesis, a pre-trained model is a speech synthesizer which hasbeen trained to synthesize the voice of another speaker. Furthermore, we can use a multi-speakerpre-trained model, which has been trained using speech recordings of multiple speakers, so itshould contain general knowledge about human speech.This paper describes how the number of speakers used to train a pre-trained model affectsthe quality of the final synthetic speech. We used a single-speaker model as well as two multispeakermodels for fine-tuning and we compared the obtained models in a listening test.

Druh

O - Ostatní výsledky

CEP obor

—

OECD FORD obor

20205 - Automation and control systems

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2024

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů