High Quality Streaming Speech Synthesis with Low, Sentence-Length-Independent Latency

Autor:	Ellinas, Nikolaos, Vamvoukakis, Georgios, Markopoulos, Konstantinos, Chalamandaris, Aimilios, Maniati, Georgia, Kakoulidis, Panos, Raptis, Spyros, Sung, June Sig, Park, Hyoungmin, Tsiakoulis, Pirros
Rok vydání:	2021
Předmět:	Computer Science - Sound Computer Science - Computation and Language Computer Science - Machine Learning Electrical Engineering and Systems Science - Audio and Speech Processing
Druh dokumentu:	Working Paper
DOI:	10.21437/Interspeech.2020-2464
Popis:	This paper presents an end-to-end text-to-speech system with low latency on a CPU, suitable for real-time applications. The system is composed of an autoregressive attention-based sequence-to-sequence acoustic model and the LPCNet vocoder for waveform generation. An acoustic model architecture that adopts modules from both the Tacotron 1 and 2 models is proposed, while stability is ensured by using a recently proposed purely location-based attention mechanism, suitable for arbitrary sentence length generation. During inference, the decoder is unrolled and acoustic feature generation is performed in a streaming manner, allowing for a nearly constant latency which is independent from the sentence length. Experimental results show that the acoustic model can produce feature sequences with minimal latency about 31 times faster than real-time on a computer CPU and 6.5 times on a mobile CPU, enabling it to meet the conditions required for real-time applications on both devices. The full end-to-end system can generate almost natural quality speech, which is verified by listening tests. Comment: Proceedings of INTERSPEECH 2020
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2111.09052 Zobrazit plný text záznamu View this record from Arxiv