We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.AI

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Sound

Title: TGAVC: Improving Autoencoder Voice Conversion with Text-Guided and Adversarial Training

Abstract: Non-parallel many-to-many voice conversion remains an interesting but challenging speech processing task. Recently, AutoVC, a conditional autoencoder based method, achieved excellent conversion results by disentangling the speaker identity and the speech content using information-constraining bottlenecks. However, due to the pure autoencoder training method, it is difficult to evaluate the separation effect of content and speaker identity. In this paper, a novel voice conversion framework, named $\boldsymbol T$ext $\boldsymbol G$uided $\boldsymbol A$utoVC(TGAVC), is proposed to more effectively separate content and timbre from speech, where an expected content embedding produced based on the text transcriptions is designed to guide the extraction of voice content. In addition, the adversarial training is applied to eliminate the speaker identity information in the estimated content embedding extracted from speech. Under the guidance of the expected content embedding and the adversarial training, the content encoder is trained to extract speaker-independent content embedding from speech. Experiments on AIShell-3 dataset show that the proposed model outperforms AutoVC in terms of naturalness and similarity of converted speech.
Comments: ASRU 6 pages
Subjects: Sound (cs.SD); Artificial Intelligence (cs.AI); Audio and Speech Processing (eess.AS)
Journal reference: 2021 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU), 2021, pp. 938-945
DOI: 10.1109/ASRU51503.2021.9688088
Cite as: arXiv:2208.04035 [cs.SD]
  (or arXiv:2208.04035v1 [cs.SD] for this version)

Submission history

From: Huaizhen Tang [view email]
[v1] Mon, 8 Aug 2022 10:33:36 GMT (303kb,D)

Link back to: arXiv, form interface, contact.