# Predicting Sounds from Video on Greatest Hits

Source: [Predicting Sounds from Video on Greatest Hits](<https://www.sota2.com/research/sota/predicting-sounds-from-video-on-greatest-hits>)

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

Data updated: 2026-02-26T05:48:47.578Z

Evaluation of models on the Greatest Hits dataset for the task of predicting sounds generated from video input, measured using metrics like Loudness Error/Correlation and Spectral Centroid Error/Correlation.

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 | Predicting Sounds from Video |
| Dataset (including any stated split) | Greatest Hits |
| Metric | Loudness Error |
| Unit | SCALAR |
| Direction | Lower 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

- [Loudness Error](<https://www.sota2.com/research/sota/predicting-sounds-from-video-on-greatest-hits.md?metric=Loudness%20Error>): Lower is better; unit: SCALAR.

- [Loudness Correlation (r)](<https://www.sota2.com/research/sota/predicting-sounds-from-video-on-greatest-hits.md?metric=Loudness%20Correlation%20(r)>): Higher is better; unit: SCALAR.

- [Spectral Centroid Error](<https://www.sota2.com/research/sota/predicting-sounds-from-video-on-greatest-hits.md?metric=Spectral%20Centroid%20Error>): Lower is better; unit: SCALAR.

- [Spectral Centroid Correlation (r)](<https://www.sota2.com/research/sota/predicting-sounds-from-video-on-greatest-hits.md?metric=Spectral%20Centroid%20Correlation%20(r)>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Loudness Error | Result date | Report paper |
| --- | --- | --- | --- |
| Full system of \[37\] | 0.21 | Not recorded | 68fa634e1e209131ce9fdf18 |
| Supervised tessellation | 0.24 | Not recorded | 68fa634e1e209131ce9fdf18 |
| Unsupervised tessellation | 0.26 | Not recorded | 68fa634e1e209131ce9fdf18 |
| Local tessellation | 0.27 | Not recorded | 68fa634e1e209131ce9fdf18 |
| Appearance matching | 0.35 | Not recorded | 68fa634e1e209131ce9fdf18 |

### Full system of \[37\]

Loudness Error: 0.21. Result ID: 698befb0ecace47ef55a6262.

[Reporting paper](<https://www.sota2.com/research/paper/68fa634e1e209131ce9fdf18.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa627d1e209131ce9f9904.md>)

Recorded evaluation configuration: {"Architecture":"CNN + LSTM"}. This record does not assert independent reproduction.

### Supervised tessellation

Loudness Error: 0.24. Result ID: 698befb0ecace47ef55a6266.

[Reporting paper](<https://www.sota2.com/research/paper/68fa634e1e209131ce9fdf18.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa634e1e209131ce9fdf18.md>)

[Implementation](<https://www.sota2.com/github.com/dot27/temporal-tessellation>)

Recorded evaluation configuration: {"Backbone":"VGG-19","Visual Representation":"RNN-FV pooled"}. This record does not assert independent reproduction.

### Unsupervised tessellation

Loudness Error: 0.26. Result ID: 698befb0ecace47ef55a6265.

[Reporting paper](<https://www.sota2.com/research/paper/68fa634e1e209131ce9fdf18.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa634e1e209131ce9fdf18.md>)

[Implementation](<https://www.sota2.com/github.com/dot27/temporal-tessellation>)

Recorded evaluation configuration: {"Backbone":"VGG-19","Visual Representation":"RNN-FV pooled"}. This record does not assert independent reproduction.

### Local tessellation

Loudness Error: 0.27. Result ID: 698befb0ecace47ef55a6264.

[Reporting paper](<https://www.sota2.com/research/paper/68fa634e1e209131ce9fdf18.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa634e1e209131ce9fdf18.md>)

[Implementation](<https://www.sota2.com/github.com/dot27/temporal-tessellation>)

Recorded evaluation configuration: {"Backbone":"VGG-19","Visual Representation":"RNN-FV pooled"}. This record does not assert independent reproduction.

### Appearance matching

Loudness Error: 0.35. Result ID: 698befb0ecace47ef55a6263.

[Reporting paper](<https://www.sota2.com/research/paper/68fa634e1e209131ce9fdf18.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.