Paper
Iterate to Differentiate: Enhancing Discriminability and Reliability in Zero-Shot TTS Evaluation
Authors
Shengfan Shen, Di Wu, Xingchen Song, Dinghao Zhou, Liumeng Xue, Meng Meng, Jian Luan, Shuai Wang
Abstract
Reliable evaluation of modern zero-shot text-to-speech (TTS) models remains challenging. Subjective tests are costly and hard to reproduce, while objective metrics often saturate, failing to distinguish SOTA systems. To address this, we propose Iterate to Differentiate (I2D), an evaluation framework that recursively synthesizes speech using the model's own outputs as references. Higher-quality models exhibit greater resilience to the distributional shift induced by iterative synthesis, resulting in slower performance degradation. I2D exploits this differential degradation to amplify performance gaps and reveal robustness. By aggregating objective metrics across iterations, I2D improves discriminability and alignment with human judgments, increasing system-level SRCC from 0.118 to 0.464 for UTMOSv2. Experiments on 11 models across Chinese, English, and emotion datasets demonstrate that I2D enables more reliable automated evaluation for zero-shot TTS.
Metadata
Related papers
Fractal universe and quantum gravity made simple
Fabio Briscese, Gianluca Calcagni • 2026-03-25
POLY-SIM: Polyglot Speaker Identification with Missing Modality Grand Challenge 2026 Evaluation Plan
Marta Moscati, Muhammad Saad Saeed, Marina Zanoni, Mubashir Noman, Rohan Kuma... • 2026-03-25
LensWalk: Agentic Video Understanding by Planning How You See in Videos
Keliang Li, Yansong Li, Hongze Shen, Mengdi Liu, Hong Chang, Shiguang Shan • 2026-03-25
Orientation Reconstruction of Proteins using Coulomb Explosions
Tomas André, Alfredo Bellisario, Nicusor Timneanu, Carl Caleman • 2026-03-25
The role of spatial context and multitask learning in the detection of organic and conventional farming systems based on Sentinel-2 time series
Jan Hemmerling, Marcel Schwieder, Philippe Rufin, Leon-Friedrich Thomas, Mire... • 2026-03-25
Raw Data (Debug)
{
"raw_xml": "<entry>\n <id>http://arxiv.org/abs/2603.24430v1</id>\n <title>Iterate to Differentiate: Enhancing Discriminability and Reliability in Zero-Shot TTS Evaluation</title>\n <updated>2026-03-25T15:36:43Z</updated>\n <link href='https://arxiv.org/abs/2603.24430v1' rel='alternate' type='text/html'/>\n <link href='https://arxiv.org/pdf/2603.24430v1' rel='related' title='pdf' type='application/pdf'/>\n <summary>Reliable evaluation of modern zero-shot text-to-speech (TTS) models remains challenging. Subjective tests are costly and hard to reproduce, while objective metrics often saturate, failing to distinguish SOTA systems. To address this, we propose Iterate to Differentiate (I2D), an evaluation framework that recursively synthesizes speech using the model's own outputs as references. Higher-quality models exhibit greater resilience to the distributional shift induced by iterative synthesis, resulting in slower performance degradation. I2D exploits this differential degradation to amplify performance gaps and reveal robustness. By aggregating objective metrics across iterations, I2D improves discriminability and alignment with human judgments, increasing system-level SRCC from 0.118 to 0.464 for UTMOSv2. Experiments on 11 models across Chinese, English, and emotion datasets demonstrate that I2D enables more reliable automated evaluation for zero-shot TTS.</summary>\n <category scheme='http://arxiv.org/schemas/atom' term='cs.SD'/>\n <published>2026-03-25T15:36:43Z</published>\n <arxiv:comment>submitted to Interspeech 2026, under review</arxiv:comment>\n <arxiv:primary_category term='cs.SD'/>\n <author>\n <name>Shengfan Shen</name>\n </author>\n <author>\n <name>Di Wu</name>\n </author>\n <author>\n <name>Xingchen Song</name>\n </author>\n <author>\n <name>Dinghao Zhou</name>\n </author>\n <author>\n <name>Liumeng Xue</name>\n </author>\n <author>\n <name>Meng Meng</name>\n </author>\n <author>\n <name>Jian Luan</name>\n </author>\n <author>\n <name>Shuai Wang</name>\n </author>\n </entry>"
}