Saved in:
Bibliographic Details
Main Authors: Giraldo, Jose, Peiró-Lilja, Alex, Zevallos, Rodolfo, España-Bonet, Cristina
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2602.05770
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908815977349120
author Giraldo, Jose
Peiró-Lilja, Alex
Zevallos, Rodolfo
España-Bonet, Cristina
author_facet Giraldo, Jose
Peiró-Lilja, Alex
Zevallos, Rodolfo
España-Bonet, Cristina
contents We evaluate two non-autoregressive architectures, StyleTTS2 and F5-TTS, to address the spontaneous nature of in-the-wild speech. Our models utilize flexible duration modeling to improve prosodic naturalness. To handle acoustic noise, we implement a multi-stage enhancement pipeline using the Sidon model, which significantly outperforms standard Demucs in signal quality. Experimental results show that finetuning enhanced audios yields superior robustness, achieving up to 4.21 UTMOS and 3.47 DNSMOS. Furthermore, we analyze the impact of reference prompt quality and length on zero-shot synthesis performance, demonstrating the effectiveness of our approach for realistic speech generation.
format Preprint
id arxiv_https___arxiv_org_abs_2602_05770
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Zero-Shot TTS With Enhanced Audio Prompts: Bsc Submission For The 2026 Wildspoof Challenge TTS Track
Giraldo, Jose
Peiró-Lilja, Alex
Zevallos, Rodolfo
España-Bonet, Cristina
Audio and Speech Processing
We evaluate two non-autoregressive architectures, StyleTTS2 and F5-TTS, to address the spontaneous nature of in-the-wild speech. Our models utilize flexible duration modeling to improve prosodic naturalness. To handle acoustic noise, we implement a multi-stage enhancement pipeline using the Sidon model, which significantly outperforms standard Demucs in signal quality. Experimental results show that finetuning enhanced audios yields superior robustness, achieving up to 4.21 UTMOS and 3.47 DNSMOS. Furthermore, we analyze the impact of reference prompt quality and length on zero-shot synthesis performance, demonstrating the effectiveness of our approach for realistic speech generation.
title Zero-Shot TTS With Enhanced Audio Prompts: Bsc Submission For The 2026 Wildspoof Challenge TTS Track
topic Audio and Speech Processing
url https://arxiv.org/abs/2602.05770