# Environment QA on LibriTTS-R clean (test)

Source: [Environment QA on LibriTTS-R clean (test)](<https://www.sota2.com/research/sota/environment-qa-on-libritts-r-clean-test>)

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

Data updated: 2026-05-15T09:48:00.604Z

Evaluation of environmental robustness (noise and reverb) for speech processing models using accuracy metrics on the clean test split of the LibriTTS-R 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 | Environment QA |
| Dataset (including any stated split) | LibriTTS-R clean (test) |
| Metric | Accuracy (Noise) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 4 of 4 |
| 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

- [Accuracy (Noise)](<https://www.sota2.com/research/sota/environment-qa-on-libritts-r-clean-test.md?metric=Accuracy%20(Noise)>): Higher is better; unit: PERCENT.

- [Accuracy (Reverb)](<https://www.sota2.com/research/sota/environment-qa-on-libritts-r-clean-test.md?metric=Accuracy%20(Reverb)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy (Noise) | Result date | Report paper |
| --- | --- | --- | --- |
| SpeakerLLM-Base | 52.7 | Not recorded | 6a06a8c4dbc836b87d82a307 |
| Qwen3.0-Omni-30B | 32.7 | Not recorded | 6a06a8c4dbc836b87d82a307 |
| Audio Flamingo3 | 21.2 | Not recorded | 6a06a8c4dbc836b87d82a307 |
| Qwen2.5-Omni-7B | 20.4 | Not recorded | 6a06a8c4dbc836b87d82a307 |

### SpeakerLLM-Base

Accuracy (Noise): 52.7. Result ID: 6a06ebba47f34ce924ef09b5.

[Reporting paper](<https://www.sota2.com/research/paper/6a06a8c4dbc836b87d82a307.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a06a8c4dbc836b87d82a307.md>)

Recorded evaluation configuration: {"LLM Backbone":"Qwen2.5-1.5B"}. This record does not assert independent reproduction.

### Qwen3.0-Omni-30B

Accuracy (Noise): 32.7. Result ID: 6a06ebba47f34ce924ef09b3.

[Reporting paper](<https://www.sota2.com/research/paper/6a06a8c4dbc836b87d82a307.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b881e209131cea88f3d.md>)

Recorded evaluation configuration: {"LLM Backbone":"Qwen3-30B-A3B"}. This record does not assert independent reproduction.

### Audio Flamingo3

Accuracy (Noise): 21.2. Result ID: 6a06ebba47f34ce924ef09b4.

[Reporting paper](<https://www.sota2.com/research/paper/6a06a8c4dbc836b87d82a307.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a4d1e209131cea82504.md>)

Recorded evaluation configuration: {"LLM Backbone":"Qwen2.5-7B"}. This record does not assert independent reproduction.

### Qwen2.5-Omni-7B

Accuracy (Noise): 20.4. Result ID: 6a06ebba47f34ce924ef09b2.

[Reporting paper](<https://www.sota2.com/research/paper/6a06a8c4dbc836b87d82a307.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71861e209131cea51c5b.md>)

[Implementation](<https://github.com/QwenLM/Qwen-Audio>)

Recorded evaluation configuration: {"LLM Backbone":"Qwen2.5-7B"}. This record does not assert independent reproduction.

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