# Cross-speaker style transfer on ESD (test)

Source: [Cross-speaker style transfer on ESD (test)](<https://www.sota2.com/research/sota/cross-speaker-style-transfer-on-esd-test>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-03-24T17:30:28.624Z

Evaluation of Cross-speaker Style Transfer performance on the test split of the ESD dataset, utilizing subjective (nMOS, eMOS, sMOS) and objective metrics (UTMOS, WER, SECS, EECS).

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Cross-speaker style transfer |
| Dataset (including any stated split) | ESD (test) |
| Metric | nMOS |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 5 of 5 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [nMOS](<https://www.sota2.com/research/sota/cross-speaker-style-transfer-on-esd-test.md?metric=nMOS>): Higher is better; unit: SCALAR.

- [eMOS](<https://www.sota2.com/research/sota/cross-speaker-style-transfer-on-esd-test.md?metric=eMOS>): Higher is better; unit: SCALAR.

- [sMOS](<https://www.sota2.com/research/sota/cross-speaker-style-transfer-on-esd-test.md?metric=sMOS>): Higher is better; unit: SCALAR.

- [UTMOS](<https://www.sota2.com/research/sota/cross-speaker-style-transfer-on-esd-test.md?metric=UTMOS>): Higher is better; unit: SCALAR.

- [WER](<https://www.sota2.com/research/sota/cross-speaker-style-transfer-on-esd-test.md?metric=WER>): Lower is better; unit: PERCENT.

- [SECS](<https://www.sota2.com/research/sota/cross-speaker-style-transfer-on-esd-test.md?metric=SECS>): Higher is better; unit: PERCENT.

- [EECS](<https://www.sota2.com/research/sota/cross-speaker-style-transfer-on-esd-test.md?metric=EECS>): Higher is better; unit: PERCENT.

## Recorded results

| Method | nMOS | Result date | Report paper |
| --- | --- | --- | --- |
| GT | 3.638 | Not recorded | 69c21ad80636e5bf4afc4db4 |
| E3-VITS | 2.763 | Not recorded | 69c21ad80636e5bf4afc4db4 |
| SelfTTS | 2.746 | Not recorded | 69c21ad80636e5bf4afc4db4 |
| VECL | 2.707 | Not recorded | 69c21ad80636e5bf4afc4db4 |
| SelfTTS w/o Self-Aug. | 2.228 | Not recorded | 69c21ad80636e5bf4afc4db4 |

### GT

nMOS: 3.638. Result ID: 69c2ca1a47f34ce924aca3fc.

[Reporting paper](<https://www.sota2.com/research/paper/69c21ad80636e5bf4afc4db4.md>)

Recorded evaluation configuration: {"Speaker Encoder":"-","Emotion Encoder":"-"}. This record does not assert independent reproduction.

### E3-VITS

nMOS: 2.763. Result ID: 69c2ca1a47f34ce924aca3fd.

[Reporting paper](<https://www.sota2.com/research/paper/69c21ad80636e5bf4afc4db4.md>)

Recorded evaluation configuration: {"Speaker Encoder":"Lookup","Emotion Encoder":"StyleSpeech"}. This record does not assert independent reproduction.

### SelfTTS

nMOS: 2.746. Result ID: 69c2ca1a47f34ce924aca400.

[Reporting paper](<https://www.sota2.com/research/paper/69c21ad80636e5bf4afc4db4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69c21ad80636e5bf4afc4db4.md>)

Recorded evaluation configuration: {"Speaker Encoder":"RE","Emotion Encoder":"RE"}. This record does not assert independent reproduction.

### VECL

nMOS: 2.707. Result ID: 69c2ca1a47f34ce924aca3fe.

[Reporting paper](<https://www.sota2.com/research/paper/69c21ad80636e5bf4afc4db4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa739c1e209131cea5d960.md>)

Recorded evaluation configuration: {"Speaker Encoder":"HSAP","Emotion Encoder":"CNN-based"}. This record does not assert independent reproduction.

### SelfTTS w/o Self-Aug.

nMOS: 2.228. Result ID: 69c2ca1a47f34ce924aca3ff.

[Reporting paper](<https://www.sota2.com/research/paper/69c21ad80636e5bf4afc4db4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69c21ad80636e5bf4afc4db4.md>)

Recorded evaluation configuration: {"Speaker Encoder":"RE","Emotion Encoder":"RE"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.