# Gesture-to-Speech Synthesis on PATS Same Text (test)

Source: [Gesture-to-Speech Synthesis on PATS Same Text (test)](<https://www.sota2.com/research/sota/gesture-to-speech-synthesis-on-pats-same-text-test>)

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

Data updated: 2026-03-23T11:28:27.618Z

Evaluation of models performing Gesture-to-Speech Synthesis using the test split of the PATS Same Text dataset, measured by metrics including WER, CER, UTMOS, and Gesture Offset.

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 | Gesture-to-Speech Synthesis |
| Dataset (including any stated split) | PATS Same Text (test) |
| Metric | Gesture Offset |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 6 of 6 |
| 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

- [Gesture Offset](<https://www.sota2.com/research/sota/gesture-to-speech-synthesis-on-pats-same-text-test.md?metric=Gesture%20Offset>): Lower is better; unit: SCALAR.

- [Mutual Information](<https://www.sota2.com/research/sota/gesture-to-speech-synthesis-on-pats-same-text-test.md?metric=Mutual%20Information>): Higher is better; unit: SCALAR.

- [WER](<https://www.sota2.com/research/sota/gesture-to-speech-synthesis-on-pats-same-text-test.md?metric=WER>): Lower is better; unit: PERCENT.

- [CER](<https://www.sota2.com/research/sota/gesture-to-speech-synthesis-on-pats-same-text-test.md?metric=CER>): Lower is better; unit: PERCENT.

- [UTMOS](<https://www.sota2.com/research/sota/gesture-to-speech-synthesis-on-pats-same-text-test.md?metric=UTMOS>): Higher is better; unit: SCALAR.

- [WVMOS](<https://www.sota2.com/research/sota/gesture-to-speech-synthesis-on-pats-same-text-test.md?metric=WVMOS>): Higher is better; unit: SCALAR.

- [AutoPCP](<https://www.sota2.com/research/sota/gesture-to-speech-synthesis-on-pats-same-text-test.md?metric=AutoPCP>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Gesture Offset | Result date | Report paper |
| --- | --- | --- | --- |
| Gesture2Speech | 0.9471 | Not recorded | 69c0c8e50636e5bf4afc3a0e |
| Gesture2Speech | 0.9794 | Not recorded | 69c0c8e50636e5bf4afc3a0e |
| Ground Truth | 1.0198 | Not recorded | 69c0c8e50636e5bf4afc3a0e |
| Gesture2Speech | 1.0386 | Not recorded | 69c0c8e50636e5bf4afc3a0e |
| Gesture2Speech | 1.2008 | Not recorded | 69c0c8e50636e5bf4afc3a0e |
| Gesture2Speech | 1.8656 | Not recorded | 69c0c8e50636e5bf4afc3a0e |

### Gesture2Speech

Gesture Offset: 0.9471. Result ID: 69c123d247f34ce924a99d60.

[Reporting paper](<https://www.sota2.com/research/paper/69c0c8e50636e5bf4afc3a0e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69c0c8e50636e5bf4afc3a0e.md>)

Recorded evaluation configuration: {"variant":"multimodal MoE"}. This record does not assert independent reproduction.

### Gesture2Speech

Gesture Offset: 0.9794. Result ID: 69c123d247f34ce924a99d5e.

[Reporting paper](<https://www.sota2.com/research/paper/69c0c8e50636e5bf4afc3a0e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69c0c8e50636e5bf4afc3a0e.md>)

Recorded evaluation configuration: {"variant":"unimodal MoE"}. This record does not assert independent reproduction.

### Ground Truth

Gesture Offset: 1.0198. Result ID: 69c123d247f34ce924a99d5b.

[Reporting paper](<https://www.sota2.com/research/paper/69c0c8e50636e5bf4afc3a0e.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Gesture2Speech

Gesture Offset: 1.0386. Result ID: 69c123d247f34ce924a99d5c.

[Reporting paper](<https://www.sota2.com/research/paper/69c0c8e50636e5bf4afc3a0e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69c0c8e50636e5bf4afc3a0e.md>)

Recorded evaluation configuration: {"variant":"XTTS V2"}. This record does not assert independent reproduction.

### Gesture2Speech

Gesture Offset: 1.2008. Result ID: 69c123d247f34ce924a99d5f.

[Reporting paper](<https://www.sota2.com/research/paper/69c0c8e50636e5bf4afc3a0e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69c0c8e50636e5bf4afc3a0e.md>)

Recorded evaluation configuration: {"variant":"H-MoE"}. This record does not assert independent reproduction.

### Gesture2Speech

Gesture Offset: 1.8656. Result ID: 69c123d247f34ce924a99d5d.

[Reporting paper](<https://www.sota2.com/research/paper/69c0c8e50636e5bf4afc3a0e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69c0c8e50636e5bf4afc3a0e.md>)

Recorded evaluation configuration: {"variant":"GPT-SoVITS"}. This record does not assert independent reproduction.

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