# Speech-to-Portrait on AVSpeech (test)

Source: [Speech-to-Portrait on AVSpeech (test)](<https://www.sota2.com/research/sota/speech-to-portrait-on-avspeech-test>)

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

Data updated: 2026-03-04T12:14:08.343Z

Evaluation of Speech-to-Portrait generation models on the test split of the AVSpeech dataset, using metrics such as L1, L2, cosine similarity, and demographic accuracy (gender/age).

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-to-Portrait |
| Dataset (including any stated split) | AVSpeech (test) |
| Metric | L1 Error |
| 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

- [L1 Error](<https://www.sota2.com/research/sota/speech-to-portrait-on-avspeech-test.md?metric=L1%20Error>): Lower is better; unit: SCALAR.

- [L2 Error](<https://www.sota2.com/research/sota/speech-to-portrait-on-avspeech-test.md?metric=L2%20Error>): Lower is better; unit: SCALAR.

- [Cosine Similarity](<https://www.sota2.com/research/sota/speech-to-portrait-on-avspeech-test.md?metric=Cosine%20Similarity>): Higher is better; unit: SCALAR.

- [Gender Accuracy](<https://www.sota2.com/research/sota/speech-to-portrait-on-avspeech-test.md?metric=Gender%20Accuracy>): Higher is better; unit: PERCENT.

- [Age Accuracy](<https://www.sota2.com/research/sota/speech-to-portrait-on-avspeech-test.md?metric=Age%20Accuracy>): Higher is better; unit: PERCENT.

- [Rank-1 Score](<https://www.sota2.com/research/sota/speech-to-portrait-on-avspeech-test.md?metric=Rank-1%20Score>): Higher is better; unit: SCALAR.

- [Rank-2 Score](<https://www.sota2.com/research/sota/speech-to-portrait-on-avspeech-test.md?metric=Rank-2%20Score>): Higher is better; unit: SCALAR.

- [Rank-5 Score](<https://www.sota2.com/research/sota/speech-to-portrait-on-avspeech-test.md?metric=Rank-5%20Score>): Higher is better; unit: SCALAR.

## Recorded results

| Method | L1 Error | Result date | Report paper |
| --- | --- | --- | --- |
| SCLDM | 31.26 | Not recorded | 690836f03ef7ccd74e5e5345 |
| Kato et al. | 46.35 | Not recorded | 690836f03ef7ccd74e5e5345 |
| Choi et al. | 60.26 | Not recorded | 690836f03ef7ccd74e5e5345 |
| Speech2Face | 67.18 | Not recorded | 690836f03ef7ccd74e5e5345 |
| SF2F | 89.31 | Not recorded | 690836f03ef7ccd74e5e5345 |
| Wav2Pix | 144.72 | Not recorded | 690836f03ef7ccd74e5e5345 |

### SCLDM

L1 Error: 31.26. Result ID: 69a821fc6dbb7075b94e33fd.

[Reporting paper](<https://www.sota2.com/research/paper/690836f03ef7ccd74e5e5345.md>)

[Method origin paper](<https://www.sota2.com/research/paper/690836f03ef7ccd74e5e5345.md>)

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

### Kato et al.

L1 Error: 46.35. Result ID: 69a821fc6dbb7075b94e33fc.

[Reporting paper](<https://www.sota2.com/research/paper/690836f03ef7ccd74e5e5345.md>)

Recorded evaluation configuration: {"Year":"2023"}. This record does not assert independent reproduction.

### Choi et al.

L1 Error: 60.26. Result ID: 69a821fc6dbb7075b94e33fa.

[Reporting paper](<https://www.sota2.com/research/paper/690836f03ef7ccd74e5e5345.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa675c1e209131cea15d97.md>)

Recorded evaluation configuration: {"Year":"2019"}. This record does not assert independent reproduction.

### Speech2Face

L1 Error: 67.18. Result ID: 69a821fc6dbb7075b94e33f9.

[Reporting paper](<https://www.sota2.com/research/paper/690836f03ef7ccd74e5e5345.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65c31e209131cea0c431.md>)

Recorded evaluation configuration: {"Year":"2019"}. This record does not assert independent reproduction.

### SF2F

L1 Error: 89.31. Result ID: 69a821fc6dbb7075b94e33fb.

[Reporting paper](<https://www.sota2.com/research/paper/690836f03ef7ccd74e5e5345.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67b71e209131cea180b0.md>)

Recorded evaluation configuration: {"Year":"2022"}. This record does not assert independent reproduction.

### Wav2Pix

L1 Error: 144.72. Result ID: 69a821fc6dbb7075b94e33f8.

[Reporting paper](<https://www.sota2.com/research/paper/690836f03ef7ccd74e5e5345.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65831e209131cea0aad6.md>)

Recorded evaluation configuration: {"Year":"2019"}. This record does not assert independent reproduction.

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