| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Keyword Spotting | Speech Commands V2 | Accuracy98.8 | 61 | |
| Audio recognition | Speech Commands V2 | Accuracy98.9 | 43 | |
| Audio Classification | Speech Commands (test) | Accuracy98.4 | 43 | |
| Audio Classification | Speech Commands V2 (test) | Accuracy98.9 | 35 | |
| 35-way Speech Classification | Speech Commands 16kHz 35-way (test) | Accuracy96.82 | 32 | |
| 35-way Speech Classification | Speech Commands 8kHz 35-way (test) | Accuracy95.05 | 28 | |
| Audio Classification | Speech Commands Spectrogram Mini (train) | Training Loss0.219 | 24 | |
| Audio Classification | Speech Commands Spectrogram Mini (val) | Accuracy83.3 | 24 | |
| Keyword Spotting | Speech Commands KS1 v1 | Accuracy98.8 | 24 | |
| Keyword Spotting | Speech Commands KS2 v2 | Accuracy98.9 | 23 | |
| Speech Command Recognition | RAW Speech Commands (SC) Full 35 Labels (val) | Accuracy96.78 | 22 | |
| Speech Classification | Speech Commands MFCC (test) | Accuracy95.3 | 16 | |
| Classification | Speech Commands Raw (SC_raw) (test) | Accuracy98.32 | 15 | |
| Speech Recognition | Speech Commands Mini | Accuracy93.4 | 13 | |
| Audio Classification | Speech Commands (SC) Unprocessed signals (RAW) | Accuracy98.32 | 13 | |
| Audio Classification | Speech Commands | Accuracy97.8 | 11 | |
| Audio Classification | Speech Commands (SC) MFCC Standard pre-processed | Accuracy95.3 | 8 | |
| Speech Command Recognition | Speech Commands 16kHz SC10 (val) | Accuracy98.51 | 7 | |
| Keyword Spotting | Speech Commands V1 | Accuracy97.3 | 6 | |
| Audio Classification | Speech Commands (SPC-1) v1 | Accuracy97.4 | 6 | |
| Speech Command Recognition | Speech Commands 8kHz SC10 (val) | Accuracy96.3 | 6 | |
| Speech classification | Speech Commands (val) | Error Rate2.22 | 6 | |
| 10-class classification | Speech Commands (SC) 10-class | Accuracy97.5 | 5 | |
| Speech Classification | Speech Commands Raw 1 → 1/2 (test) | Accuracy88.66 | 5 | |
| Speech Classification | Speech Commands Raw 1 → 1 (test) | Accuracy95.87 | 5 |