Redirecting to original paper in 30 seconds...
Click below to go immediately or wait for automatic redirect
📄 Abstract
Abstract: We analyze the syntactic sensitivity of Text-to-Speech (TTS) systems using
methods inspired by psycholinguistic research. Specifically, we focus on the
generation of intonational phrase boundaries, which can often be predicted by
identifying syntactic boundaries within a sentence. We find that TTS systems
struggle to accurately generate intonational phrase boundaries in sentences
where syntactic boundaries are ambiguous (e.g., garden path sentences or
sentences with attachment ambiguity). In these cases, systems need superficial
cues such as commas to place boundaries at the correct positions. In contrast,
for sentences with simpler syntactic structures, we find that systems do
incorporate syntactic cues beyond surface markers. Finally, we finetune models
on sentences without commas at the syntactic boundary positions, encouraging
them to focus on more subtle linguistic cues. Our findings indicate that this
leads to more distinct intonation patterns that better reflect the underlying
structure.
Authors (3)
Charlotte Pouw
Afra Alishahi
Willem Zuidema
Key Contributions
This paper analyzes the syntactic sensitivity of TTS systems regarding intonational phrasing, revealing that they struggle with syntactically ambiguous sentences and often rely on superficial cues like commas. It demonstrates that fine-tuning models on sentences without commas can improve their focus on subtle linguistic cues for more distinct intonation.
Business Value
Leads to more natural and human-like synthetic voices, enhancing user experience in voice assistants, audiobooks, and other speech-based applications.