# Speech Enhancement on TCD-TIMIT 3 Speakers (test)

Source: [Speech Enhancement on TCD-TIMIT 3 Speakers (test)](<https://www.sota2.com/research/sota/speech-enhancement-on-tcd-timit-3-speakers-test>)

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

Data updated: 2026-02-26T06:02:50.246Z

Evaluation of Speech Enhancement performance on the test split of the TCD-TIMIT 3 Speakers dataset, using metrics like SDR, PESQ, and ViSQOL.

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 Enhancement |
| Dataset (including any stated split) | TCD-TIMIT 3 Speakers (test) |
| Metric | SDR |
| 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

- [SDR](<https://www.sota2.com/research/sota/speech-enhancement-on-tcd-timit-3-speakers-test.md?metric=SDR>): Higher is better; unit: SCALAR.

- [PESQ](<https://www.sota2.com/research/sota/speech-enhancement-on-tcd-timit-3-speakers-test.md?metric=PESQ>): Higher is better; unit: SCALAR.

- [ViSQOL](<https://www.sota2.com/research/sota/speech-enhancement-on-tcd-timit-3-speakers-test.md?metric=ViSQOL>): Higher is better; unit: SCALAR.

## Recorded results

| Method | SDR | Result date | Report paper |
| --- | --- | --- | --- |
| AV c-ref | 5.37 | Not recorded | 68fa69371e209131cea20e84 |
| VL2M\_ref | 5.27 | Not recorded | 68fa69371e209131cea20e84 |
| AV concat | 5.15 | Not recorded | 68fa69371e209131cea20e84 |
| VL2M | -0.51 | Not recorded | 68fa69371e209131cea20e84 |
| Noisy | -3.42 | Not recorded | 68fa69371e209131cea20e84 |

### AV c-ref

SDR: 5.37. Result ID: 698d9e588b517eb43322b5e8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa69371e209131cea20e84.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69371e209131cea20e84.md>)

[Implementation](<https://goo.gl/3h1NgE>)

Recorded evaluation configuration: {"type":"Speaker-independent","features":"Landmark motion vectors","talkers":3}. This record does not assert independent reproduction.

### VL2M\_ref

SDR: 5.27. Result ID: 698d9e588b517eb43322b5e6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa69371e209131cea20e84.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69371e209131cea20e84.md>)

[Implementation](<https://goo.gl/3h1NgE>)

Recorded evaluation configuration: {"type":"Speaker-independent","features":"Landmark motion vectors","talkers":3}. This record does not assert independent reproduction.

### AV concat

SDR: 5.15. Result ID: 698d9e588b517eb43322b5e7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa69371e209131cea20e84.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69371e209131cea20e84.md>)

[Implementation](<https://goo.gl/3h1NgE>)

Recorded evaluation configuration: {"type":"Speaker-independent","features":"Landmark motion vectors","talkers":3}. This record does not assert independent reproduction.

### VL2M

SDR: -0.51. Result ID: 698d9e588b517eb43322b5e5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa69371e209131cea20e84.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69371e209131cea20e84.md>)

[Implementation](<https://goo.gl/3h1NgE>)

Recorded evaluation configuration: {"type":"Speaker-independent","features":"Landmark motion vectors","talkers":3}. This record does not assert independent reproduction.

### Noisy

SDR: -3.42. Result ID: 698d9e588b517eb43322b5e4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa69371e209131cea20e84.md>)

Recorded evaluation configuration: {"type":"Input mixed-speech signal","talkers":3}. This record does not assert independent reproduction.

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