Incremental Text-to-Speech Synthesis with Prefix-to-Prefix Framework
This addresses latency issues in real-time applications like simultaneous translation and assistive technologies, representing a novel method for a known bottleneck.
The paper tackles the problem of latency in text-to-speech synthesis, including computational and input latencies, by proposing the first neural incremental TTS approach based on a prefix-to-prefix framework, achieving O(1) latency instead of O(n).
Text-to-speech synthesis (TTS) has witnessed rapid progress in recent years, where neural methods became capable of producing audios with high naturalness. However, these efforts still suffer from two types of latencies: (a) the {\em computational latency} (synthesizing time), which grows linearly with the sentence length even with parallel approaches, and (b) the {\em input latency} in scenarios where the input text is incrementally generated (such as in simultaneous translation, dialog generation, and assistive technologies). To reduce these latencies, we devise the first neural incremental TTS approach based on the recently proposed prefix-to-prefix framework. We synthesize speech in an online fashion, playing a segment of audio while generating the next, resulting in an $O(1)$ rather than $O(n)$ latency.