# Articulatory-to-Acoustic Inversion (AAI) on EMA dataset unseen speakers (test)

Source: [Articulatory-to-Acoustic Inversion (AAI) on EMA dataset unseen speakers (test)](<https://www.sota2.com/research/sota/articulatory-to-acoustic-inversion-aai-on-ema-dataset-unseen-speakers-test>)

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

Data updated: 2026-07-03T06:51:02.310Z

Evaluation of Articulatory-to-Acoustic Inversion (AAI) performance on the test split of the EMA dataset, focusing on unseen speakers. Metrics include Correlation Coefficient (CC) and Root Mean Square Error (RMSE).

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 | Articulatory-to-Acoustic Inversion (AAI) |
| Dataset (including any stated split) | EMA dataset unseen speakers (test) |
| Metric | CC |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 24 of 24 |
| 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

- [CC](<https://www.sota2.com/research/sota/articulatory-to-acoustic-inversion-aai-on-ema-dataset-unseen-speakers-test.md?metric=CC>): Higher is better; unit: SCALAR.

- [RMSE](<https://www.sota2.com/research/sota/articulatory-to-acoustic-inversion-aai-on-ema-dataset-unseen-speakers-test.md?metric=RMSE>): Lower is better; unit: SCALAR.

## Recorded results

| Method | CC | Result date | Report paper |
| --- | --- | --- | --- |
| ACP-T | 0.7818 | Not recorded | 6a474208143ebdcacacc5bdf |
| ACP-T | 0.781 | Not recorded | 6a474208143ebdcacacc5bdf |
| Baseline | 0.7777 | Not recorded | 6a474208143ebdcacacc5bdf |
| ACP-T | 0.7755 | Not recorded | 6a474208143ebdcacacc5bdf |
| Baseline | 0.7717 | Not recorded | 6a474208143ebdcacacc5bdf |
| ACP-T | 0.7689 | Not recorded | 6a474208143ebdcacacc5bdf |
| Baseline | 0.7664 | Not recorded | 6a474208143ebdcacacc5bdf |
| ACP-T | 0.7653 | Not recorded | 6a474208143ebdcacacc5bdf |
| ACP-T | 0.7621 | Not recorded | 6a474208143ebdcacacc5bdf |
| ACP-T | 0.7616 | Not recorded | 6a474208143ebdcacacc5bdf |
| Baseline | 0.7563 | Not recorded | 6a474208143ebdcacacc5bdf |
| ACP-T | 0.7562 | Not recorded | 6a474208143ebdcacacc5bdf |
| ACP-T | 0.7561 | Not recorded | 6a474208143ebdcacacc5bdf |
| Baseline | 0.754 | Not recorded | 6a474208143ebdcacacc5bdf |
| Baseline | 0.7488 | Not recorded | 6a474208143ebdcacacc5bdf |
| Baseline | 0.7488 | Not recorded | 6a474208143ebdcacacc5bdf |
| ACP-T | 0.7469 | Not recorded | 6a474208143ebdcacacc5bdf |
| ACP-T | 0.7399 | Not recorded | 6a474208143ebdcacacc5bdf |
| Baseline | 0.7396 | Not recorded | 6a474208143ebdcacacc5bdf |
| Baseline | 0.7325 | Not recorded | 6a474208143ebdcacacc5bdf |
| Baseline | 0.7265 | Not recorded | 6a474208143ebdcacacc5bdf |
| ACP-T | 0.7259 | Not recorded | 6a474208143ebdcacacc5bdf |
| Baseline | 0.6991 | Not recorded | 6a474208143ebdcacacc5bdf |
| Baseline | 0.6687 | Not recorded | 6a474208143ebdcacacc5bdf |

### ACP-T

CC: 0.7818. Result ID: 6a475bd660b88bb3399234f9.

[Reporting paper](<https://www.sota2.com/research/paper/6a474208143ebdcacacc5bdf.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a474208143ebdcacacc5bdf.md>)

Recorded evaluation configuration: {"Inputs":"TERA","EMA data used for training (%)":"75"}. This record does not assert independent reproduction.

### ACP-T

CC: 0.781. Result ID: 6a475bd660b88bb3399234fa.

[Reporting paper](<https://www.sota2.com/research/paper/6a474208143ebdcacacc5bdf.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a474208143ebdcacacc5bdf.md>)

Recorded evaluation configuration: {"Inputs":"TERA","EMA data used for training (%)":"100"}. This record does not assert independent reproduction.

### Baseline

CC: 0.7777. Result ID: 6a475bd660b88bb3399234f3.

[Reporting paper](<https://www.sota2.com/research/paper/6a474208143ebdcacacc5bdf.md>)

Recorded evaluation configuration: {"Inputs":"TERA","EMA data used for training (%)":"75"}. This record does not assert independent reproduction.

### ACP-T

CC: 0.7755. Result ID: 6a475bd660b88bb3399234f8.

[Reporting paper](<https://www.sota2.com/research/paper/6a474208143ebdcacacc5bdf.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a474208143ebdcacacc5bdf.md>)

Recorded evaluation configuration: {"Inputs":"TERA","EMA data used for training (%)":"50"}. This record does not assert independent reproduction.

### Baseline

CC: 0.7717. Result ID: 6a475bd660b88bb3399234f4.

[Reporting paper](<https://www.sota2.com/research/paper/6a474208143ebdcacacc5bdf.md>)

Recorded evaluation configuration: {"Inputs":"TERA","EMA data used for training (%)":"100"}. This record does not assert independent reproduction.

### ACP-T

CC: 0.7689. Result ID: 6a475bd660b88bb3399234ee.

[Reporting paper](<https://www.sota2.com/research/paper/6a474208143ebdcacacc5bdf.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a474208143ebdcacacc5bdf.md>)

Recorded evaluation configuration: {"Inputs":"MFCC","EMA data used for training (%)":"100"}. This record does not assert independent reproduction.

### Baseline

CC: 0.7664. Result ID: 6a475bd660b88bb3399234f2.

[Reporting paper](<https://www.sota2.com/research/paper/6a474208143ebdcacacc5bdf.md>)

Recorded evaluation configuration: {"Inputs":"TERA","EMA data used for training (%)":"50"}. This record does not assert independent reproduction.

### ACP-T

CC: 0.7653. Result ID: 6a475bd660b88bb3399234ed.

[Reporting paper](<https://www.sota2.com/research/paper/6a474208143ebdcacacc5bdf.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a474208143ebdcacacc5bdf.md>)

Recorded evaluation configuration: {"Inputs":"MFCC","EMA data used for training (%)":"75"}. This record does not assert independent reproduction.

### ACP-T

CC: 0.7621. Result ID: 6a475bd660b88bb3399234f7.

[Reporting paper](<https://www.sota2.com/research/paper/6a474208143ebdcacacc5bdf.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a474208143ebdcacacc5bdf.md>)

Recorded evaluation configuration: {"Inputs":"TERA","EMA data used for training (%)":"25"}. This record does not assert independent reproduction.

### ACP-T

CC: 0.7616. Result ID: 6a475bd660b88bb3399234ec.

[Reporting paper](<https://www.sota2.com/research/paper/6a474208143ebdcacacc5bdf.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a474208143ebdcacacc5bdf.md>)

Recorded evaluation configuration: {"Inputs":"MFCC","EMA data used for training (%)":"50"}. This record does not assert independent reproduction.

### Baseline

CC: 0.7563. Result ID: 6a475bd660b88bb3399234e8.

[Reporting paper](<https://www.sota2.com/research/paper/6a474208143ebdcacacc5bdf.md>)

Recorded evaluation configuration: {"Inputs":"MFCC","EMA data used for training (%)":"100"}. This record does not assert independent reproduction.

### ACP-T

CC: 0.7562. Result ID: 6a475bd660b88bb3399234f6.

[Reporting paper](<https://www.sota2.com/research/paper/6a474208143ebdcacacc5bdf.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a474208143ebdcacacc5bdf.md>)

Recorded evaluation configuration: {"Inputs":"TERA","EMA data used for training (%)":"12.5"}. This record does not assert independent reproduction.

### ACP-T

CC: 0.7561. Result ID: 6a475bd660b88bb3399234f5.

[Reporting paper](<https://www.sota2.com/research/paper/6a474208143ebdcacacc5bdf.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a474208143ebdcacacc5bdf.md>)

Recorded evaluation configuration: {"Inputs":"TERA","EMA data used for training (%)":"6.25"}. This record does not assert independent reproduction.

### Baseline

CC: 0.754. Result ID: 6a475bd660b88bb3399234f1.

[Reporting paper](<https://www.sota2.com/research/paper/6a474208143ebdcacacc5bdf.md>)

Recorded evaluation configuration: {"Inputs":"TERA","EMA data used for training (%)":"25"}. This record does not assert independent reproduction.

### Baseline

CC: 0.7488. Result ID: 6a475bd660b88bb3399234e6.

[Reporting paper](<https://www.sota2.com/research/paper/6a474208143ebdcacacc5bdf.md>)

Recorded evaluation configuration: {"Inputs":"MFCC","EMA data used for training (%)":"50"}. This record does not assert independent reproduction.

### Baseline

CC: 0.7488. Result ID: 6a475bd660b88bb3399234e7.

[Reporting paper](<https://www.sota2.com/research/paper/6a474208143ebdcacacc5bdf.md>)

Recorded evaluation configuration: {"Inputs":"MFCC","EMA data used for training (%)":"75"}. This record does not assert independent reproduction.

### ACP-T

CC: 0.7469. Result ID: 6a475bd660b88bb3399234eb.

[Reporting paper](<https://www.sota2.com/research/paper/6a474208143ebdcacacc5bdf.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a474208143ebdcacacc5bdf.md>)

Recorded evaluation configuration: {"Inputs":"MFCC","EMA data used for training (%)":"25"}. This record does not assert independent reproduction.

### ACP-T

CC: 0.7399. Result ID: 6a475bd660b88bb3399234ea.

[Reporting paper](<https://www.sota2.com/research/paper/6a474208143ebdcacacc5bdf.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a474208143ebdcacacc5bdf.md>)

Recorded evaluation configuration: {"Inputs":"MFCC","EMA data used for training (%)":"12.5"}. This record does not assert independent reproduction.

### Baseline

CC: 0.7396. Result ID: 6a475bd660b88bb3399234f0.

[Reporting paper](<https://www.sota2.com/research/paper/6a474208143ebdcacacc5bdf.md>)

Recorded evaluation configuration: {"Inputs":"TERA","EMA data used for training (%)":"12.5"}. This record does not assert independent reproduction.

### Baseline

CC: 0.7325. Result ID: 6a475bd660b88bb3399234ef.

[Reporting paper](<https://www.sota2.com/research/paper/6a474208143ebdcacacc5bdf.md>)

Recorded evaluation configuration: {"Inputs":"TERA","EMA data used for training (%)":"6.25"}. This record does not assert independent reproduction.

### Baseline

CC: 0.7265. Result ID: 6a475bd660b88bb3399234e5.

[Reporting paper](<https://www.sota2.com/research/paper/6a474208143ebdcacacc5bdf.md>)

Recorded evaluation configuration: {"Inputs":"MFCC","EMA data used for training (%)":"25"}. This record does not assert independent reproduction.

### ACP-T

CC: 0.7259. Result ID: 6a475bd660b88bb3399234e9.

[Reporting paper](<https://www.sota2.com/research/paper/6a474208143ebdcacacc5bdf.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a474208143ebdcacacc5bdf.md>)

Recorded evaluation configuration: {"Inputs":"MFCC","EMA data used for training (%)":"6.25"}. This record does not assert independent reproduction.

### Baseline

CC: 0.6991. Result ID: 6a475bd660b88bb3399234e4.

[Reporting paper](<https://www.sota2.com/research/paper/6a474208143ebdcacacc5bdf.md>)

Recorded evaluation configuration: {"Inputs":"MFCC","EMA data used for training (%)":"12.5"}. This record does not assert independent reproduction.

### Baseline

CC: 0.6687. Result ID: 6a475bd660b88bb3399234e3.

[Reporting paper](<https://www.sota2.com/research/paper/6a474208143ebdcacacc5bdf.md>)

Recorded evaluation configuration: {"Inputs":"MFCC","EMA data used for training (%)":"6.25"}. This record does not assert independent reproduction.

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