# Music Transcription on MAESTRO (Onset FMS)

Source: [Music Transcription on MAESTRO (Onset FMS)](<https://www.sota2.com/research/sota/music-transcription-on-maestro-onset-fms>)

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

Data updated: 2026-02-26T05:54:16.265Z

Evaluation of Music Transcription performance on the MAESTRO dataset, utilizing Onset FMS and Onset w/ Offset FMS metrics.

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 | Music transcription |
| Dataset (including any stated split) | MAESTRO |
| Metric | Onset FMS |
| 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

- [Onset FMS](<https://www.sota2.com/research/sota/music-transcription-on-maestro-onset-fms.md?metric=Onset%20FMS>): Higher is better; unit: SCALAR.

- [Onset w/ Offset FMS](<https://www.sota2.com/research/sota/music-transcription-on-maestro-onset-fms.md?metric=Onset%20w%2F%20Offset%20FMS>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Onset FMS | Result date | Report paper |
| --- | --- | --- | --- |
| HEAR CREPE | 0.391 | Not recorded | 68fa6e3e1e209131cea3ea1a |
| ABT AudioNTT | 0.04801 | Not recorded | 68fa6e3e1e209131cea3ea1a |
| HEAR wav2vec2.0 | 0.0328 | Not recorded | 68fa6e3e1e209131cea3ea1a |
| ABT ViTC-B | 0.0263 | Not recorded | 68fa6e3e1e209131cea3ea1a |
| HEAR BYOL-A\* | 0.0028 | Not recorded | 68fa6e3e1e209131cea3ea1a |

### HEAR CREPE

Onset FMS: 0.391. Result ID: 699009993a6c314849fda789.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e3e1e209131cea3ea1a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa64111e209131cea02831.md>)

Recorded evaluation configuration: {"Backbone":"CREPE"}. This record does not assert independent reproduction.

### ABT AudioNTT

Onset FMS: 0.04801. Result ID: 699009993a6c314849fda78c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e3e1e209131cea3ea1a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e3e1e209131cea3ea1a.md>)

[Implementation](<https://github.com/jonahanton/SSL_audio>)

Recorded evaluation configuration: {"Backbone":"AudioNTT"}. This record does not assert independent reproduction.

### HEAR wav2vec2.0

Onset FMS: 0.0328. Result ID: 699009993a6c314849fda78a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e3e1e209131cea3ea1a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68891e209131cea1cc36.md>)

[Implementation](<https://github.com/pytorch/fairseq>)

Recorded evaluation configuration: {"Backbone":"wav2vec2.0"}. This record does not assert independent reproduction.

### ABT ViTC-B

Onset FMS: 0.0263. Result ID: 699009993a6c314849fda78d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e3e1e209131cea3ea1a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e3e1e209131cea3ea1a.md>)

[Implementation](<https://github.com/jonahanton/SSL_audio>)

Recorded evaluation configuration: {"Backbone":"ViTC-B","Patch size":"16 x 8"}. This record does not assert independent reproduction.

### HEAR BYOL-A\*

Onset FMS: 0.0028. Result ID: 699009993a6c314849fda78b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e3e1e209131cea3ea1a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6dba1e209131cea3b9bf.md>)

[Implementation](<https://github.com/GasserElbanna/serab-byols>)

Recorded evaluation configuration: {"Backbone":"BYOL-A\*"}. This record does not assert independent reproduction.

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