© 2025 Sony Interactive Entertainment Europe Limited (SIEE)
Tous les contenus, titres de jeux, noms commerciaux et/ou habillage commercial, marques déposées, illustrations et images associées sont des marques déposées et/ou la propriété en droit d'auteur de leurs propriétaires respectifs. Tous droits réservés. Plus d'informations
N° Registre du Commerce : 122587 - MARRAKECH.
Taxe professionnelle : 45307423.
Numéro d’affiliation à la CNSS : 4008035.
Identifiant Fiscal (IF) : 51721336.
Identifiant commun de l’entreprise (ICE) : 002999298000083.
We evaluate VALL-E on LibriSpeech [Panayotov et al., 2015] and VCTK [Veaux et al., 2016]datasets, where all test speakers are unseen in the training corpus.
VALL-E significantly outperforms the state-of-the-art zero-shot TTS system ( YourTTS (2021) ) in terms of speech naturalness and speaker similarity, with + comparative mean option score (CMOS) and + similarity mean option score (SMOS) improvement on LibriSpeech.
VALL-E also beats the baseline on VCTK with+ SMOS and + CMOS improvements.
It even achieves a + CMOS score against ground truth, showing the synthesized speech of unseen speakers is as natural as human recordings on VCTK.
Moreover, the qualitative analysis shows that VALL-E is able to synthesize diverse outputs with the same text and target speaker, which could benefit pseudo-data creation for the speech recognition task.
We also find that VALL-E could keep the acoustic environment (., reverberation) and emotion (. anger) of the acoustic prompt.
VALL-E 2 is purely a research project. Currently, we have no plans to incorporate VALL-E 2 into a product or expand access to the public. VALL-E 2 could synthesize speech that maintains speaker identity and could be used for educational learning, entertainment, journalistic, self-authored content, accessibility features, interactive voice response systems, translation, chatbot, and so on. While VALL-E 2 can speak in a voice like the voice talent, the similarity, and naturalness depend on the length and quality of the speech prompt, the background noise, as well as other factors. It may carry potential risks in the misuse of the model, such as spoofing voice identification or impersonating a specific speaker. We conducted the experiments under the assumption that the user agrees to be the target speaker in speech synthesis. If the model is generalized to unseen speakers in the real world, it should include a protocol to ensure that the speaker approves the use of their voice and a synthesized speech detection model. If you suspect that VALL-E 2 is being used in a manner that is abusive or illegal or infringes on your rights or the rights of other people, you can report it at the Report Abuse Portal.
Tags:
Playstation,
5,
видео,
обзор,
игр,
