# Binaural Audio Synthesis on Binaural Speech Dataset (test)

Source: [Binaural Audio Synthesis on Binaural Speech Dataset (test)](<https://www.sota2.com/research/sota/binaural-audio-synthesis-on-binaural-speech-dataset-test>)

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

Data updated: 2026-02-26T06:08:44.241Z

Evaluation of Binaural Audio Synthesis performance using the test split of the Binaural Speech Dataset, measured via Wave L2, Amplitude L2, Phase L2, PESQ, and MRSTFT.

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 | Binaural Audio Synthesis |
| Dataset (including any stated split) | Binaural Speech Dataset (test) |
| Metric | Wave L2 |
| Unit | SCALAR |
| Direction | Lower 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

- [Wave L2](<https://www.sota2.com/research/sota/binaural-audio-synthesis-on-binaural-speech-dataset-test.md?metric=Wave%20L2>): Lower is better; unit: SCALAR.

- [Amplitude L2](<https://www.sota2.com/research/sota/binaural-audio-synthesis-on-binaural-speech-dataset-test.md?metric=Amplitude%20L2>): Lower is better; unit: SCALAR.

- [Phase L2](<https://www.sota2.com/research/sota/binaural-audio-synthesis-on-binaural-speech-dataset-test.md?metric=Phase%20L2>): Lower is better; unit: SCALAR.

- [PESQ](<https://www.sota2.com/research/sota/binaural-audio-synthesis-on-binaural-speech-dataset-test.md?metric=PESQ>): Higher is better; unit: SCALAR.

- [MRSTFT](<https://www.sota2.com/research/sota/binaural-audio-synthesis-on-binaural-speech-dataset-test.md?metric=MRSTFT>): Lower is better; unit: SCALAR.

## Recorded results

| Method | Wave L2 | Result date | Report paper |
| --- | --- | --- | --- |
| BinauralGrad | 0.128 | Not recorded | 68fa6e061e209131cea3d7dd |
| WarpNet | 0.157 | Not recorded | 68fa6e061e209131cea3d7dd |
| WaveNet | 0.179 | Not recorded | 68fa6e061e209131cea3d7dd |
| DSP | 1.543 | Not recorded | 68fa6e061e209131cea3d7dd |

### BinauralGrad

Wave L2: 0.128. Result ID: 698ff5093a6c314849fd1170.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e061e209131cea3d7dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e061e209131cea3d7dd.md>)

[Implementation](<https://github.com/microsoft/NeuralSpeech/tree/master/BinauralGrad>)

Recorded evaluation configuration: {"Two-stage framework":"true"}. This record does not assert independent reproduction.

### WarpNet

Wave L2: 0.157. Result ID: 698ff5093a6c314849fd116f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e061e209131cea3d7dd.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### WaveNet

Wave L2: 0.179. Result ID: 698ff5093a6c314849fd116e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e061e209131cea3d7dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa62bf1e209131ce9fb1ba.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### DSP

Wave L2: 1.543. Result ID: 698ff5093a6c314849fd116d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e061e209131cea3d7dd.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

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