# Speech Naturalness Assessment on CVSS-T

Source: [Speech Naturalness Assessment on CVSS-T](<https://www.sota2.com/research/sota/speech-naturalness-assessment-on-cvss-t>)

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

Data updated: 2026-02-26T06:07:51.387Z

Evaluation of speech naturalness using the UTMOS metric on the CVSS-T dataset.

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 | Speech Naturalness Assessment |
| Dataset (including any stated split) | CVSS-T |
| Metric | UTMOS |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 10 of 10 |
| 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

- [UTMOS](<https://www.sota2.com/research/sota/speech-naturalness-assessment-on-cvss-t.md?metric=UTMOS>): Higher is better; unit: SCALAR.

## Recorded results

| Method | UTMOS | Result date | Report paper |
| --- | --- | --- | --- |
| Tri-Task Learning | 3.67 | Not recorded | 6976f4e6a2b28e761d87c0c7 |
| Tri-Task Learning | 3.66 | Not recorded | 6976f4e6a2b28e761d87c0c7 |
| Vanilla | 3.64 | Not recorded | 6976f4e6a2b28e761d87c0c7 |
| Vanilla | 3.64 | Not recorded | 6976f4e6a2b28e761d87c0c7 |
| Tri-Task Learning | 3.63 | Not recorded | 6976f4e6a2b28e761d87c0c7 |
| Chain of Modality | 3.61 | Not recorded | 6976f4e6a2b28e761d87c0c7 |
| Vanilla | 3.61 | Not recorded | 6976f4e6a2b28e761d87c0c7 |
| Chain of Modality | 3.61 | Not recorded | 6976f4e6a2b28e761d87c0c7 |
| Chain of Modality | 3.6 | Not recorded | 6976f4e6a2b28e761d87c0c7 |
| S2T+TTS | 3.08 | Not recorded | 6976f4e6a2b28e761d87c0c7 |

### Tri-Task Learning

UTMOS: 3.67. Result ID: 699671de5698665ec4b90977.

[Reporting paper](<https://www.sota2.com/research/paper/6976f4e6a2b28e761d87c0c7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6976f4e6a2b28e761d87c0c7.md>)

Recorded evaluation configuration: {"System":"End-to-end","Optimization Stage":"After DPO","Back-translation Evaluation Metric":"BLEU"}. This record does not assert independent reproduction.

### Tri-Task Learning

UTMOS: 3.66. Result ID: 699671de5698665ec4b90974.

[Reporting paper](<https://www.sota2.com/research/paper/6976f4e6a2b28e761d87c0c7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6976f4e6a2b28e761d87c0c7.md>)

Recorded evaluation configuration: {"System":"End-to-end","Optimization Stage":"After DPO","Back-translation Evaluation Metric":"METEOR"}. This record does not assert independent reproduction.

### Vanilla

UTMOS: 3.64. Result ID: 699671de5698665ec4b90970.

[Reporting paper](<https://www.sota2.com/research/paper/6976f4e6a2b28e761d87c0c7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6976f4e6a2b28e761d87c0c7.md>)

Recorded evaluation configuration: {"System":"End-to-end","Optimization Stage":"Before DPO"}. This record does not assert independent reproduction.

### Vanilla

UTMOS: 3.64. Result ID: 699671de5698665ec4b90973.

[Reporting paper](<https://www.sota2.com/research/paper/6976f4e6a2b28e761d87c0c7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6976f4e6a2b28e761d87c0c7.md>)

Recorded evaluation configuration: {"System":"End-to-end","Optimization Stage":"After DPO","Back-translation Evaluation Metric":"METEOR"}. This record does not assert independent reproduction.

### Tri-Task Learning

UTMOS: 3.63. Result ID: 699671de5698665ec4b90971.

[Reporting paper](<https://www.sota2.com/research/paper/6976f4e6a2b28e761d87c0c7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6976f4e6a2b28e761d87c0c7.md>)

Recorded evaluation configuration: {"System":"End-to-end","Optimization Stage":"Before DPO"}. This record does not assert independent reproduction.

### Chain of Modality

UTMOS: 3.61. Result ID: 699671de5698665ec4b90975.

[Reporting paper](<https://www.sota2.com/research/paper/6976f4e6a2b28e761d87c0c7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6976f4e6a2b28e761d87c0c7.md>)

Recorded evaluation configuration: {"System":"End-to-end","Optimization Stage":"After DPO","Back-translation Evaluation Metric":"METEOR"}. This record does not assert independent reproduction.

### Vanilla

UTMOS: 3.61. Result ID: 699671de5698665ec4b90976.

[Reporting paper](<https://www.sota2.com/research/paper/6976f4e6a2b28e761d87c0c7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6976f4e6a2b28e761d87c0c7.md>)

Recorded evaluation configuration: {"System":"End-to-end","Optimization Stage":"After DPO","Back-translation Evaluation Metric":"BLEU"}. This record does not assert independent reproduction.

### Chain of Modality

UTMOS: 3.61. Result ID: 699671de5698665ec4b90978.

[Reporting paper](<https://www.sota2.com/research/paper/6976f4e6a2b28e761d87c0c7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6976f4e6a2b28e761d87c0c7.md>)

Recorded evaluation configuration: {"System":"End-to-end","Optimization Stage":"After DPO","Back-translation Evaluation Metric":"BLEU"}. This record does not assert independent reproduction.

### Chain of Modality

UTMOS: 3.6. Result ID: 699671de5698665ec4b90972.

[Reporting paper](<https://www.sota2.com/research/paper/6976f4e6a2b28e761d87c0c7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6976f4e6a2b28e761d87c0c7.md>)

Recorded evaluation configuration: {"System":"End-to-end","Optimization Stage":"Before DPO"}. This record does not assert independent reproduction.

### S2T+TTS

UTMOS: 3.08. Result ID: 699671de5698665ec4b9096f.

[Reporting paper](<https://www.sota2.com/research/paper/6976f4e6a2b28e761d87c0c7.md>)

Recorded evaluation configuration: {"System":"Cascaded"}. This record does not assert independent reproduction.

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