# Paralinguistic-aware dialogue on ParaS2SBench Real (Human Evaluation)

Source: [Paralinguistic-aware dialogue on ParaS2SBench Real (Human Evaluation)](<https://www.sota2.com/research/sota/paralinguistic-aware-dialogue-on-paras2sbench-real-human-evaluation>)

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

Data updated: 2026-03-09T06:59:19.258Z

Evaluation of models on the ParaS2SBench Real dataset focusing on paralinguistic-aware dialogue capabilities, typically involving human evaluation and reporting aggregated scores across components like IEMOCAP and MELD.

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 | Paralinguistic-aware dialogue |
| Dataset (including any stated split) | ParaS2SBench Real (Human Evaluation) |
| Metric | IEMOCAP Score |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 8 of 8 |
| 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

- [IEMOCAP Score](<https://www.sota2.com/research/sota/paralinguistic-aware-dialogue-on-paras2sbench-real-human-evaluation.md?metric=IEMOCAP%20Score>): Higher is better; unit: SCALAR.

- [MELD Score](<https://www.sota2.com/research/sota/paralinguistic-aware-dialogue-on-paras2sbench-real-human-evaluation.md?metric=MELD%20Score>): Higher is better; unit: SCALAR.

- [Average Score](<https://www.sota2.com/research/sota/paralinguistic-aware-dialogue-on-paras2sbench-real-human-evaluation.md?metric=Average%20Score>): Higher is better; unit: SCALAR.

## Recorded results

| Method | IEMOCAP Score | Result date | Report paper |
| --- | --- | --- | --- |
| GPT-TTS | 4.911 | Not recorded | 691566140052cac1dc65d8c4 |
| Kimi-Audio GRPO | 4.336 | Not recorded | 691566140052cac1dc65d8c4 |
| Kimi-Audio SFT | 4.212 | Not recorded | 691566140052cac1dc65d8c4 |
| GPT-4o voice mode | 3.862 | Not recorded | 691566140052cac1dc65d8c4 |
| Qwen2.5 Omni | 3.713 | Not recorded | 691566140052cac1dc65d8c4 |
| Whisper-GPT-TTS | 3.601 | Not recorded | 691566140052cac1dc65d8c4 |
| GLM 4 | 3.521 | Not recorded | 691566140052cac1dc65d8c4 |
| Kimi-Audio | 2.231 | Not recorded | 691566140052cac1dc65d8c4 |

### GPT-TTS

IEMOCAP Score: 4.911. Result ID: 69ae6e916dbb7075b955f62d.

[Reporting paper](<https://www.sota2.com/research/paper/691566140052cac1dc65d8c4.md>)

Recorded evaluation configuration: {"Category":"Topline"}. This record does not assert independent reproduction.

### Kimi-Audio GRPO

IEMOCAP Score: 4.336. Result ID: 69ae6e916dbb7075b955f62c.

[Reporting paper](<https://www.sota2.com/research/paper/691566140052cac1dc65d8c4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/691566140052cac1dc65d8c4.md>)

Recorded evaluation configuration: {"Category":"Ours"}. This record does not assert independent reproduction.

### Kimi-Audio SFT

IEMOCAP Score: 4.212. Result ID: 69ae6e916dbb7075b955f62b.

[Reporting paper](<https://www.sota2.com/research/paper/691566140052cac1dc65d8c4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/691566140052cac1dc65d8c4.md>)

Recorded evaluation configuration: {"Category":"Ours"}. This record does not assert independent reproduction.

### GPT-4o voice mode

IEMOCAP Score: 3.862. Result ID: 69ae6e916dbb7075b955f627.

[Reporting paper](<https://www.sota2.com/research/paper/691566140052cac1dc65d8c4.md>)

Recorded evaluation configuration: {"Category":"Closed Source"}. This record does not assert independent reproduction.

### Qwen2.5 Omni

IEMOCAP Score: 3.713. Result ID: 69ae6e916dbb7075b955f628.

[Reporting paper](<https://www.sota2.com/research/paper/691566140052cac1dc65d8c4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77a21e209131cea7482c.md>)

Recorded evaluation configuration: {"Category":"Open Source"}. This record does not assert independent reproduction.

### Whisper-GPT-TTS

IEMOCAP Score: 3.601. Result ID: 69ae6e916dbb7075b955f626.

[Reporting paper](<https://www.sota2.com/research/paper/691566140052cac1dc65d8c4.md>)

Recorded evaluation configuration: {"Category":"Baseline"}. This record does not assert independent reproduction.

### GLM 4

IEMOCAP Score: 3.521. Result ID: 69ae6e916dbb7075b955f629.

[Reporting paper](<https://www.sota2.com/research/paper/691566140052cac1dc65d8c4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75f91e209131cea6abba.md>)

Recorded evaluation configuration: {"Category":"Open Source"}. This record does not assert independent reproduction.

### Kimi-Audio

IEMOCAP Score: 2.231. Result ID: 69ae6e916dbb7075b955f62a.

[Reporting paper](<https://www.sota2.com/research/paper/691566140052cac1dc65d8c4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78431e209131cea77803.md>)

Recorded evaluation configuration: {"Category":"Open Source"}. This record does not assert independent reproduction.

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