Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386875722> ?p ?o ?g. }
Showing items 1 to 93 of
93
with 100 items per page.
- W4386875722 abstract "Recent advancements in speech synthesis have leveraged GAN-based networks like HiFi-GAN and BigVGAN to produce high-fidelity waveforms from mel-spectrograms. However, these networks are computationally expensive and parameter-heavy. iSTFTNet addresses these limitations by integrating inverse short-time Fourier transform (iSTFT) into the network, achieving both speed and parameter efficiency. In this paper, we introduce an extension to iSTFTNet, termed HiFTNet, which incorporates a harmonic-plus-noise source filter in the time-frequency domain that uses a sinusoidal source from the fundamental frequency (F0) inferred via a pre-trained F0 estimation network for fast inference speed. Subjective evaluations on LJSpeech show that our model significantly outperforms both iSTFTNet and HiFi-GAN, achieving ground-truth-level performance. HiFTNet also outperforms BigVGAN-base on LibriTTS for unseen speakers and achieves comparable performance to BigVGAN while being four times faster with only $1/6$ of the parameters. Our work sets a new benchmark for efficient, high-quality neural vocoding, paving the way for real-time applications that demand high quality speech synthesis." @default.
- W4386875722 created "2023-09-20" @default.
- W4386875722 creator A5023800090 @default.
- W4386875722 creator A5025321643 @default.
- W4386875722 creator A5033351155 @default.
- W4386875722 creator A5048984565 @default.
- W4386875722 date "2023-09-18" @default.
- W4386875722 modified "2023-09-27" @default.
- W4386875722 title "HiFTNet: A Fast High-Quality Neural Vocoder with Harmonic-plus-Noise Filter and Inverse Short Time Fourier Transform" @default.
- W4386875722 doi "https://doi.org/10.48550/arxiv.2309.09493" @default.
- W4386875722 hasPublicationYear "2023" @default.
- W4386875722 type Work @default.
- W4386875722 citedByCount "0" @default.
- W4386875722 crossrefType "posted-content" @default.
- W4386875722 hasAuthorship W4386875722A5023800090 @default.
- W4386875722 hasAuthorship W4386875722A5025321643 @default.
- W4386875722 hasAuthorship W4386875722A5033351155 @default.
- W4386875722 hasAuthorship W4386875722A5048984565 @default.
- W4386875722 hasBestOaLocation W43868757221 @default.
- W4386875722 hasConcept C102519508 @default.
- W4386875722 hasConcept C106131492 @default.
- W4386875722 hasConcept C11413529 @default.
- W4386875722 hasConcept C115961682 @default.
- W4386875722 hasConcept C121332964 @default.
- W4386875722 hasConcept C127934551 @default.
- W4386875722 hasConcept C13280743 @default.
- W4386875722 hasConcept C134306372 @default.
- W4386875722 hasConcept C154945302 @default.
- W4386875722 hasConcept C166386157 @default.
- W4386875722 hasConcept C185798385 @default.
- W4386875722 hasConcept C197424946 @default.
- W4386875722 hasConcept C203024314 @default.
- W4386875722 hasConcept C205649164 @default.
- W4386875722 hasConcept C207467116 @default.
- W4386875722 hasConcept C24890656 @default.
- W4386875722 hasConcept C2524010 @default.
- W4386875722 hasConcept C2779530757 @default.
- W4386875722 hasConcept C2779948431 @default.
- W4386875722 hasConcept C28490314 @default.
- W4386875722 hasConcept C31972630 @default.
- W4386875722 hasConcept C33923547 @default.
- W4386875722 hasConcept C41008148 @default.
- W4386875722 hasConcept C45273575 @default.
- W4386875722 hasConcept C50644808 @default.
- W4386875722 hasConcept C554190296 @default.
- W4386875722 hasConcept C62520636 @default.
- W4386875722 hasConcept C76155785 @default.
- W4386875722 hasConcept C99498987 @default.
- W4386875722 hasConceptScore W4386875722C102519508 @default.
- W4386875722 hasConceptScore W4386875722C106131492 @default.
- W4386875722 hasConceptScore W4386875722C11413529 @default.
- W4386875722 hasConceptScore W4386875722C115961682 @default.
- W4386875722 hasConceptScore W4386875722C121332964 @default.
- W4386875722 hasConceptScore W4386875722C127934551 @default.
- W4386875722 hasConceptScore W4386875722C13280743 @default.
- W4386875722 hasConceptScore W4386875722C134306372 @default.
- W4386875722 hasConceptScore W4386875722C154945302 @default.
- W4386875722 hasConceptScore W4386875722C166386157 @default.
- W4386875722 hasConceptScore W4386875722C185798385 @default.
- W4386875722 hasConceptScore W4386875722C197424946 @default.
- W4386875722 hasConceptScore W4386875722C203024314 @default.
- W4386875722 hasConceptScore W4386875722C205649164 @default.
- W4386875722 hasConceptScore W4386875722C207467116 @default.
- W4386875722 hasConceptScore W4386875722C24890656 @default.
- W4386875722 hasConceptScore W4386875722C2524010 @default.
- W4386875722 hasConceptScore W4386875722C2779530757 @default.
- W4386875722 hasConceptScore W4386875722C2779948431 @default.
- W4386875722 hasConceptScore W4386875722C28490314 @default.
- W4386875722 hasConceptScore W4386875722C31972630 @default.
- W4386875722 hasConceptScore W4386875722C33923547 @default.
- W4386875722 hasConceptScore W4386875722C41008148 @default.
- W4386875722 hasConceptScore W4386875722C45273575 @default.
- W4386875722 hasConceptScore W4386875722C50644808 @default.
- W4386875722 hasConceptScore W4386875722C554190296 @default.
- W4386875722 hasConceptScore W4386875722C62520636 @default.
- W4386875722 hasConceptScore W4386875722C76155785 @default.
- W4386875722 hasConceptScore W4386875722C99498987 @default.
- W4386875722 hasLocation W43868757221 @default.
- W4386875722 hasOpenAccess W4386875722 @default.
- W4386875722 hasPrimaryLocation W43868757221 @default.
- W4386875722 hasRelatedWork W2061798471 @default.
- W4386875722 hasRelatedWork W2108153523 @default.
- W4386875722 hasRelatedWork W2129331087 @default.
- W4386875722 hasRelatedWork W2346761963 @default.
- W4386875722 hasRelatedWork W2802845977 @default.
- W4386875722 hasRelatedWork W2921820890 @default.
- W4386875722 hasRelatedWork W3004580690 @default.
- W4386875722 hasRelatedWork W3133517635 @default.
- W4386875722 hasRelatedWork W4296765717 @default.
- W4386875722 hasRelatedWork W4378805784 @default.
- W4386875722 isParatext "false" @default.
- W4386875722 isRetracted "false" @default.
- W4386875722 workType "article" @default.