Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

The ISCSLP 2024 Conversational Voice Clone (CoVoC) Challenge: Tasks, Results and Findings

Kangxiang Xia, Dake Guo, Jixun Yao, Liumeng Xue, Hanzhao Li, Shuai Wang, Zhao Guo, Lei Xie, Qingqing Zhang, Lei Luo, Minghui Dong, Peng Sun

Abstract

The ISCSLP 2024 Conversational Voice Clone (CoVoC) Challenge aims to benchmark and advance zero-shot spontaneous style voice cloning, particularly focusing on generating spontaneous behaviors in conversational speech. The challenge comprises two tracks: an unconstrained track without limitation on data and model usage, and a constrained track only allowing the use of constrained open-source datasets. A 100-hour high-quality conversational speech dataset is also made available with the challenge. This paper details the data, tracks, submitted systems, evaluation results, and findings.

The ISCSLP 2024 Conversational Voice Clone (CoVoC) Challenge: Tasks, Results and Findings

Abstract

The ISCSLP 2024 Conversational Voice Clone (CoVoC) Challenge aims to benchmark and advance zero-shot spontaneous style voice cloning, particularly focusing on generating spontaneous behaviors in conversational speech. The challenge comprises two tracks: an unconstrained track without limitation on data and model usage, and a constrained track only allowing the use of constrained open-source datasets. A 100-hour high-quality conversational speech dataset is also made available with the challenge. This paper details the data, tracks, submitted systems, evaluation results, and findings.

Paper Structure

This paper contains 13 sections, 1 equation, 4 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Challenge Design
  3. Track Setting
  4. Test Datasets
  5. Metrics
  6. Objective Metrics
  7. Subjective Metrics
  8. Submitted systems
  9. Results and Analysis
  10. Analysis between CER and Text Type
  11. Analysis between SIM and Prompt Duration
  12. Analysis between SSS and Prompt Style
  13. Conclusion

Figures (4)

  • Figure 1: Average CER for each team across different text sets. The redder the color, the higher the CER.
  • Figure 2: The CER results between different text lengths.
  • Figure 3: The similarity results between different duration lengths of the audio prompt.
  • Figure 4: MOS score of SSS between two types of audio prompt.