Advancing NAM-to-Speech Conversion with Novel Methods and the MultiNAM Dataset

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Shah, Neil, Karande, Shirish, Gandhi, Vineet
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915117251166208
author Shah, Neil
Karande, Shirish
Gandhi, Vineet
author_facet Shah, Neil
Karande, Shirish
Gandhi, Vineet
contents Current Non-Audible Murmur (NAM)-to-speech techniques rely on voice cloning to simulate ground-truth speech from paired whispers. However, the simulated speech often lacks intelligibility and fails to generalize well across different speakers. To address this issue, we focus on learning phoneme-level alignments from paired whispers and text and employ a Text-to-Speech (TTS) system to simulate the ground-truth. To reduce dependence on whispers, we learn phoneme alignments directly from NAMs, though the quality is constrained by the available training data. To further mitigate reliance on NAM/whisper data for ground-truth simulation, we propose incorporating the lip modality to infer speech and introduce a novel diffusion-based method that leverages recent advancements in lip-to-speech technology. Additionally, we release the MultiNAM dataset with over 7.96 hours of paired NAM, whisper, video, and text data from two speakers and benchmark all methods on this dataset. Speech samples and the dataset are available at https://diff-nam.github.io/DiffNAM/
format Preprint
id arxiv_https___arxiv_org_abs_2412_18839
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Advancing NAM-to-Speech Conversion with Novel Methods and the MultiNAM Dataset
Shah, Neil
Karande, Shirish
Gandhi, Vineet
Sound
Artificial Intelligence
Audio and Speech Processing
Current Non-Audible Murmur (NAM)-to-speech techniques rely on voice cloning to simulate ground-truth speech from paired whispers. However, the simulated speech often lacks intelligibility and fails to generalize well across different speakers. To address this issue, we focus on learning phoneme-level alignments from paired whispers and text and employ a Text-to-Speech (TTS) system to simulate the ground-truth. To reduce dependence on whispers, we learn phoneme alignments directly from NAMs, though the quality is constrained by the available training data. To further mitigate reliance on NAM/whisper data for ground-truth simulation, we propose incorporating the lip modality to infer speech and introduce a novel diffusion-based method that leverages recent advancements in lip-to-speech technology. Additionally, we release the MultiNAM dataset with over 7.96 hours of paired NAM, whisper, video, and text data from two speakers and benchmark all methods on this dataset. Speech samples and the dataset are available at https://diff-nam.github.io/DiffNAM/
title Advancing NAM-to-Speech Conversion with Novel Methods and the MultiNAM Dataset
topic Sound
Artificial Intelligence
Audio and Speech Processing
url https://arxiv.org/abs/2412.18839