No abstract

Background: Circular RNAs (circRNAs) act as competing endogenous RNAs (ceRNAs) that indirectly regulate gene expression and function by binding microRNAs (miRNAs). A growing body of evidence indicates that the ceRNA networks can be used as an effective method to investigate cancer; however, the construction and analysis of ceRNA networks, especially circRNA-miRNA-mRNA regulatory network, in different subtypes of breast cancer has not been previously performed. Methods: First, the expression profiles of circRNA, miRNA, and mRNA were downloaded from the GEO database, differentially expressed genes were obtained using GEO2R, and a ceRNA network, was constructed based on circRNA-miRNA pairs and miRNA-mRNA pairs, which was consisted of 10 circRNAs, 25 miRNAs and 39 mRNAs. Further studies of BC subtypes based on TCGA datasets were also performed to validate the effect of novel ceRNA network. Results and Discussion: Then, the related genes in the regulatory network were analyzed by GO functional annotation and KEGG pathway enrichment. The analysis showed that targeted genes were enriched in 97 GO terms and 25 KEGG pathways, respectively, involved in the molecular typing of breast cancer. Meanwhile, Kaplan-Meier analysis revealed that three key genes (MKI67, DEF8, and GFRA1) were significantly associated with BC tumor differentiation and prognosis. Conclusion: The current study provides a potential application of ceRNA network within BC subtypes, and may offer new targets for their diagnosis, therapy and prognosis.

Typical high quality text-to-speech (TTS) systems today use a two-stage architecture, with a spectrum model stage that generates spectral frames and a vocoder stage that generates the actual audio. High-quality spectrum models usually incorporate the encoder-decoder architecture with selfattention or bi-directional long short-term (BLSTM) units. While these models can produce high quality speech, they often incur O(L) increase in both latency and real-time factor (RTF) with respect to input length L. In other words, longer inputs leads to longer delay and slower synthesis speed, limiting its use in real-time applications. In this paper, we propose a multi-rate attention architecture that breaks the latency and RTF bottlenecks by computing a compact representation during encoding and recurrently generating the attention vector in a streaming manner during decoding. The proposed architecture achieves high audio quality (MOS of 4.31 compared to groundtruth 4.48), low latency, and low RTF at the same time. Meanwhile, both latency and RTF of the proposed system stay constant regardless of input lengths, making it ideal for real-time applications.

No abstract

Typical high quality text-to-speech (TTS) systems today use a two-stage architecture, with a spectrum model stage that generates spectral frames and a vocoder stage that generates the actual audio. High-quality spectrum models usually incorporate the encoder-decoder architecture with selfattention or bi-directional long short-term (BLSTM) units. While these models can produce high quality speech, they often incur O(L) increase in both latency and real-time factor (RTF) with respect to input length L. In other words, longer inputs leads to longer delay and slower synthesis speed, limiting its use in real-time applications. In this paper, we propose a multi-rate attention architecture that breaks the latency and RTF bottlenecks by computing a compact representation during encoding and recurrently generating the attention vector in a streaming manner during decoding. The proposed architecture achieves high audio quality (MOS of 4.31 compared to groundtruth 4.48), low latency, and low RTF at the same time. Meanwhile, both latency and RTF of the proposed system stay constant regardless of input lengths, making it ideal for real-time applications.