Diogo-V's picture
Duplicate from Diogo-V/wav2vec-nlr-40c5c9eb9345
d193ebd verified
| | accuracy | bits | clip | dataset | method | model | params | peak_mem_alloc_mb | peak_mem_reserv_mb | perplexity | time_ms | wer |
|---:|-----------:|-------:|-------:|:------------|:---------|:--------|:----------|--------------------:|---------------------:|-------------:|----------:|-------:|
| 0 | 0 | 4 | 0 | librispeech | comb_7 | wav2vec | nlr_t_lin | 534.743 | 1633.68 | 0 | 27.8148 | 1 |
| 1 | 0 | 5 | 0 | librispeech | comb_7 | wav2vec | nlr_t_lin | 534.743 | 1633.68 | 0 | 27.8148 | 1 |
| 2 | 0 | 6 | 0 | librispeech | comb_7 | wav2vec | nlr_t_lin | 534.743 | 1633.68 | 0 | 27.8148 | 1 |
| 3 | 0 | 7 | 0 | librispeech | comb_7 | wav2vec | nlr_t_lin | 534.743 | 1633.68 | 0 | 27.8148 | 1 |
| 4 | 0 | 8 | 0 | librispeech | comb_7 | wav2vec | nlr_t_lin | 534.743 | 1633.68 | 0 | 27.8148 | 1 |
| 5 | 0 | 4 | 0 | librispeech | comb_4 | wav2vec | nlr_t_lin | 534.743 | 1633.68 | 0 | 37.7273 | 1 |
| 6 | 0 | 5 | 0 | librispeech | comb_4 | wav2vec | nlr_t_lin | 534.743 | 1633.68 | 0 | 37.7273 | 1 |
| 7 | 0 | 6 | 0 | librispeech | comb_4 | wav2vec | nlr_t_lin | 534.743 | 1633.68 | 0 | 37.7273 | 1 |
| 8 | 0 | 7 | 0 | librispeech | comb_4 | wav2vec | nlr_t_lin | 534.743 | 1633.68 | 0 | 37.7273 | 1 |
| 9 | 0 | 8 | 0 | librispeech | comb_4 | wav2vec | nlr_t_lin | 534.743 | 1633.68 | 0 | 37.7273 | 1 |
| 10 | 0 | 4 | 0 | librispeech | comb_17 | wav2vec | nlr_t_lin | 535.632 | 1633.68 | 0 | 32.0422 | 1 |
| 11 | 0 | 5 | 0 | librispeech | comb_17 | wav2vec | nlr_t_lin | 535.632 | 1633.68 | 0 | 32.0422 | 1 |
| 12 | 0 | 6 | 0 | librispeech | comb_17 | wav2vec | nlr_t_lin | 535.632 | 1633.68 | 0 | 32.0422 | 1 |
| 13 | 0 | 7 | 0 | librispeech | comb_17 | wav2vec | nlr_t_lin | 535.632 | 1633.68 | 0 | 32.0422 | 1 |
| 14 | 0 | 8 | 0 | librispeech | comb_17 | wav2vec | nlr_t_lin | 535.632 | 1633.68 | 0 | 32.0422 | 1 |
| 15 | 0 | 4 | 0 | librispeech | comb_1 | wav2vec | nlr_t_lin | 535.632 | 1633.68 | 0 | 34.8728 | 1 |
| 16 | 0 | 5 | 0 | librispeech | comb_1 | wav2vec | nlr_t_lin | 535.632 | 1633.68 | 0 | 34.8728 | 1 |
| 17 | 0 | 4 | 0 | librispeech | comb_2 | wav2vec | nlr_t_lin | 535.632 | 1633.68 | 0 | 111.374 | 1 |
| 18 | 0 | 5 | 0 | librispeech | comb_2 | wav2vec | nlr_t_lin | 535.632 | 1633.68 | 0 | 111.374 | 1 |
| 19 | 0 | 6 | 0 | librispeech | comb_2 | wav2vec | nlr_t_lin | 535.632 | 1633.68 | 0 | 111.374 | 1 |
| 20 | 0 | 7 | 0 | librispeech | comb_2 | wav2vec | nlr_t_lin | 535.632 | 1633.68 | 0 | 111.374 | 1 |
| 21 | 0 | 8 | 0 | librispeech | comb_2 | wav2vec | nlr_t_lin | 535.632 | 1633.68 | 0 | 111.374 | 1 |
| 22 | 0 | 4 | 0 | librispeech | comb_6 | wav2vec | nlr_t_lin | 535.632 | 1633.68 | 0 | 121.667 | 1 |
| 23 | 0 | 5 | 0 | librispeech | comb_6 | wav2vec | nlr_t_lin | 535.632 | 1633.68 | 0 | 121.667 | 1 |
| 24 | 0 | 6 | 0 | librispeech | comb_6 | wav2vec | nlr_t_lin | 535.632 | 1633.68 | 0 | 121.667 | 1 |
| 25 | 0 | 7 | 0 | librispeech | comb_6 | wav2vec | nlr_t_lin | 535.632 | 1633.68 | 0 | 121.667 | 1 |
| 26 | 0 | 8 | 0 | librispeech | comb_6 | wav2vec | nlr_t_lin | 535.632 | 1633.68 | 0 | 121.667 | 1 |
| 27 | 0 | 4 | 0 | librispeech | comb_11 | wav2vec | nlr_t_lin | 368.001 | 1642.07 | 0 | 0.0494 | 1 |
| 28 | 0 | 5 | 0 | librispeech | comb_11 | wav2vec | nlr_t_lin | 368.001 | 1642.07 | 0 | 0.0494 | 1 |
| 29 | 0 | 6 | 0 | librispeech | comb_11 | wav2vec | nlr_t_lin | 368.001 | 1642.07 | 0 | 0.0494 | 1 |
| 30 | 0 | 7 | 0 | librispeech | comb_11 | wav2vec | nlr_t_lin | 368.001 | 1642.07 | 0 | 0.0494 | 1 |
| 31 | 0 | 8 | 0 | librispeech | comb_11 | wav2vec | nlr_t_lin | 368.001 | 1642.07 | 0 | 0.0494 | 1 |
| 32 | 0 | 8 | 0 | librispeech | comb_15 | wav2vec | nlr_t_lin | 658.018 | 4829.74 | 0 | 25.6234 | 0.0239 |
| 33 | 0 | 7 | 0 | librispeech | comb_15 | wav2vec | nlr_t_lin | 658.018 | 4829.74 | 0 | 26.1562 | 0.024 |
| 34 | 0 | 6 | 0 | librispeech | comb_15 | wav2vec | nlr_t_lin | 658.018 | 4829.74 | 0 | 33.5202 | 0.0235 |
| 35 | 0 | 5 | 0 | librispeech | comb_15 | wav2vec | nlr_t_lin | 658.018 | 4829.74 | 0 | 40.9004 | 0.0235 |
| 36 | 0 | 4 | 0 | librispeech | comb_15 | wav2vec | nlr_t_lin | 658.018 | 4829.74 | 0 | 107.638 | 0.0262 |
| 37 | 0 | 8 | 0 | librispeech | woq | wav2vec | fixed | 658.092 | 4831.84 | 0 | 34.0541 | 0.0239 |
| 38 | 0 | 5 | 0 | librispeech | woq | wav2vec | fixed | 658.092 | 4831.84 | 0 | 35.4665 | 0.0237 |
| 39 | 0 | 6 | 0 | librispeech | woq | wav2vec | fixed | 658.092 | 4831.84 | 0 | 38.6328 | 0.0234 |
| 40 | 0 | 4 | 0 | librispeech | woq | wav2vec | fixed | 658.092 | 4831.84 | 0 | 107.707 | 0.0269 |
| 41 | 0 | 7 | 0 | librispeech | woq | wav2vec | fixed | 658.092 | 4831.84 | 0 | 112.778 | 0.0236 |
| 42 | 0 | 8 | 0 | librispeech | comb_1 | wav2vec | nlr_t_lin | 658.907 | 4831.84 | 0 | 36.2118 | 0.0308 |
| 43 | 0 | 8 | 0 | librispeech | comb_12 | wav2vec | nlr_t_lin | 658.907 | 4831.84 | 0 | 38.6678 | 0.0239 |
| 44 | 0 | 8 | 0 | librispeech | comb_8 | wav2vec | nlr_t_lin | 658.907 | 4831.84 | 0 | 39.8882 | 0.0285 |
| 45 | 0 | 4 | 0 | librispeech | comb_8 | wav2vec | nlr_t_lin | 658.907 | 4831.84 | 0 | 40.435 | 0.0408 |
| 46 | 0 | 8 | 0 | librispeech | comb_3 | wav2vec | nlr_t_lin | 658.907 | 4831.84 | 0 | 41.651 | 0.0239 |
| 47 | 0 | 5 | 0 | librispeech | comb_5 | wav2vec | nlr_t_lin | 658.907 | 4831.84 | 0 | 43.985 | 0.0274 |
| 48 | 0 | 6 | 0 | librispeech | comb_18 | wav2vec | nlr_t_lin | 658.907 | 4831.84 | 0 | 48.3737 | 0.0235 |
| 49 | 0 | 4 | 0 | librispeech | comb_12 | wav2vec | nlr_t_lin | 658.907 | 4831.84 | 0 | 49.0833 | 0.0261 |
| 50 | 0 | 6 | 0 | librispeech | comb_3 | wav2vec | nlr_t_lin | 658.907 | 4831.84 | 0 | 49.6859 | 0.0235 |
| 51 | 0 | 5 | 0 | librispeech | comb_12 | wav2vec | nlr_t_lin | 658.907 | 4831.84 | 0 | 54.8932 | 0.0244 |
| 52 | 0 | 6 | 0 | librispeech | comb_5 | wav2vec | nlr_t_lin | 658.907 | 4831.84 | 0 | 57.3485 | 0.0272 |
| 53 | 0 | 8 | 0 | librispeech | comb_18 | wav2vec | nlr_t_lin | 658.907 | 4831.84 | 0 | 57.7205 | 0.0241 |
| 54 | 0 | 7 | 0 | librispeech | comb_8 | wav2vec | nlr_t_lin | 658.907 | 4831.84 | 0 | 59.4552 | 0.0283 |
| 55 | 0 | 4 | 0 | librispeech | comb_5 | wav2vec | nlr_t_lin | 658.907 | 4831.84 | 0 | 63.9388 | 0.0288 |
| 56 | 0 | 8 | 0 | librispeech | comb_10 | wav2vec | nlr_t_lin | 658.907 | 4831.84 | 0 | 75.6275 | 0.0236 |
| 57 | 0 | 7 | 0 | librispeech | comb_1 | wav2vec | nlr_t_lin | 658.907 | 4831.84 | 0 | 108 | 0.0283 |
| 58 | 0 | 7 | 0 | librispeech | comb_18 | wav2vec | nlr_t_lin | 658.907 | 4831.84 | 0 | 108.29 | 0.024 |
| 59 | 0 | 5 | 0 | librispeech | comb_18 | wav2vec | nlr_t_lin | 658.907 | 4831.84 | 0 | 114.197 | 0.0233 |
| 60 | 0 | 8 | 0 | librispeech | comb_5 | wav2vec | nlr_t_lin | 658.907 | 4831.84 | 0 | 115.246 | 0.0268 |
| 61 | 0 | 7 | 0 | librispeech | comb_3 | wav2vec | nlr_t_lin | 658.907 | 4831.84 | 0 | 116.667 | 0.0235 |
| 62 | 0 | 7 | 0 | librispeech | comb_12 | wav2vec | nlr_t_lin | 658.907 | 4831.84 | 0 | 118.075 | 0.0236 |
| 63 | 0 | 5 | 0 | librispeech | comb_8 | wav2vec | nlr_t_lin | 658.907 | 4831.84 | 0 | 118.338 | 0.0321 |
| 64 | 0 | 6 | 0 | librispeech | comb_1 | wav2vec | nlr_t_lin | 658.907 | 4831.84 | 0 | 120.23 | 0.7174 |
| 65 | 0 | 6 | 0 | librispeech | comb_10 | wav2vec | nlr_t_lin | 658.907 | 4831.84 | 0 | 120.698 | 0.0238 |
| 66 | 0 | 5 | 0 | librispeech | comb_10 | wav2vec | nlr_t_lin | 658.907 | 4831.84 | 0 | 122.462 | 0.027 |
| 67 | 0 | 4 | 0 | librispeech | comb_3 | wav2vec | nlr_t_lin | 658.907 | 4831.84 | 0 | 122.955 | 0.0251 |
| 68 | 0 | 6 | 0 | librispeech | comb_8 | wav2vec | nlr_t_lin | 658.907 | 4831.84 | 0 | 124.222 | 0.0285 |
| 69 | 0 | 4 | 0 | librispeech | comb_18 | wav2vec | nlr_t_lin | 658.907 | 4831.84 | 0 | 125.827 | 0.0255 |
| 70 | 0 | 7 | 0 | librispeech | comb_5 | wav2vec | nlr_t_lin | 658.907 | 4831.84 | 0 | 127.407 | 0.0259 |
| 71 | 0 | 5 | 0 | librispeech | comb_3 | wav2vec | nlr_t_lin | 658.907 | 4831.84 | 0 | 127.958 | 0.0238 |
| 72 | 0 | 4 | 0 | librispeech | comb_10 | wav2vec | nlr_t_lin | 658.907 | 4831.84 | 0 | 130.054 | 0.0435 |
| 73 | 0 | 7 | 0 | librispeech | comb_10 | wav2vec | nlr_t_lin | 658.907 | 4831.84 | 0 | 131.588 | 0.0238 |
| 74 | 0 | 6 | 0 | librispeech | comb_12 | wav2vec | nlr_t_lin | 658.907 | 4831.84 | 0 | 132.288 | 0.0235 |
| 75 | 0 | 8 | 0 | librispeech | ao_woq | wav2vec | fixed | 658.018 | 4890.56 | 0 | 34.8716 | 0.024 |
| 76 | 0 | 32 | 0 | librispeech | base | wav2vec | ---- | 1564.77 | 5620.37 | 0 | 105.862 | 0.0239 |