{ "input": "examples/en/source.wav", "seconds": 7.84, "renderer": { "soundfont_sha256": "5b85b6c2c61d10b2b91cddd41efcce7b25cd31c8271d511c73afafbef20b6fa3", "sr": 24000, "hop": 240, "nfft": 2048, "preset_name": "Grand Piano", "renderer_library_sha256": "6ef97148241a3c529ad1a528631a3610aa5ec75f3846ba5534cb494ddbe6812c", "pitches": [ 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108 ], "velocities": [ 45, 75, 105 ], "durations": [ 0.06, 0.12, 0.24 ], "tail": 0.3, "gain": 0.8, "bank": 0, "program": 0, "effects": false, "renderer_revision": 3 }, "results": { "baseline": { "spectral_convergence": 0.8311071395874023, "log_band_mae": 0.4577343463897705, "waveform_rmse": 0.038239434361457825, "peak": 0.08251953125, "notes": 457 }, "dictionary": { "spectral_convergence": 0.7595574259757996, "log_band_mae": 0.41065511107444763, "waveform_rmse": 0.04594961181282997, "peak": 0.2164306640625, "notes": 159 }, "refined": { "spectral_convergence": 0.7606934309005737, "log_band_mae": 0.3874739408493042, "waveform_rmse": 0.04269171506166458, "peak": 0.1861724853515625, "notes": 148 } }, "render_repeatability": { "spectral_convergence": 0.00038946481072343886, "log_band_loss_delta": 2.9802322387695312e-06 }, "pursuit": { "accepted": 159, "surrogate_improvements": [ 10558.4384765625, 7962.7666015625, 5715.4150390625, 4834.4609375, 4818.0126953125, 4130.3291015625, 3561.078369140625, 3054.567626953125, 2828.303955078125, 2729.42138671875, 2669.8310546875, 2600.94384765625, 2541.796630859375, 2499.138671875, 2358.1962890625, 2025.1171875, 1868.8902587890625, 1587.477783203125, 1505.8052978515625, 1470.310302734375, 1463.44091796875, 1455.779296875, 1395.0450439453125, 1251.4173583984375, 1196.8651123046875, 1175.1383056640625, 1097.8623046875, 1049.9212646484375, 1043.9747314453125, 1032.818603515625, 969.533935546875, 900.7930908203125, 889.942138671875, 868.384521484375, 818.9688720703125, 804.3232421875, 788.57861328125, 767.5830078125, 765.97119140625, 764.38671875, 726.8178100585938, 705.6483154296875, 681.6700439453125, 655.1786499023438, 632.2413940429688, 607.3171997070312, 597.638427734375, 592.790283203125, 566.4576416015625, 550.314208984375, 545.9325561523438, 516.3138427734375, 498.017333984375, 496.83941650390625, 475.308837890625, 470.1158447265625, 458.0395202636719, 437.9156494140625, 434.23394775390625, 418.92169189453125, 414.6388244628906, 409.8388671875, 406.0827331542969, 397.35064697265625, 361.15875244140625, 351.2940979003906, 350.638671875, 348.0697021484375, 344.16351318359375, 320.3564758300781, 312.5291442871094, 304.381103515625, 302.6427917480469, 298.3839416503906, 277.4814453125, 258.9993896484375, 258.8589172363281, 257.4494934082031, 256.14251708984375, 251.90460205078125, 247.6282958984375, 237.59478759765625, 218.486083984375, 206.99249267578125, 196.5882568359375, 194.52203369140625, 194.25827026367188, 192.5644989013672, 189.775390625, 186.1031494140625, 182.21157836914062, 179.97972106933594, 172.13833618164062, 168.379150390625, 163.37557983398438, 162.050537109375, 159.35293579101562, 150.99884033203125, 143.433349609375, 142.3600311279297, 133.32754516601562, 131.35986328125, 129.4129638671875, 128.5037841796875, 124.34005737304688, 120.27095031738281, 113.77438354492188, 113.61799621582031, 111.92544555664062, 106.53375244140625, 97.86294555664062, 97.17410278320312, 92.25390625, 84.85130310058594, 83.4329833984375, 83.315673828125, 79.22296142578125, 71.72164916992188, 67.56948852539062, 63.90354919433594, 63.26670837402344, 62.13752746582031, 61.413909912109375, 59.041595458984375, 58.68309020996094, 53.02793884277344, 51.157684326171875, 50.46064758300781, 49.80511474609375, 49.77580261230469, 48.7662353515625, 46.16965103149414, 44.27165222167969, 40.14201354980469, 39.092742919921875, 38.146484375, 36.064605712890625, 33.02142333984375, 30.11712646484375, 29.4931640625, 28.2784423828125, 25.933815002441406, 25.50750732421875, 23.485584259033203, 21.60456085205078, 20.85321044921875, 19.01141357421875, 18.26642608642578, 17.060089111328125, 14.386226654052734, 12.91082763671875, 12.244163513183594, 11.771728515625, 10.889877319335938, 6.941280364990234, 6.2841339111328125, 4.660541534423828, 4.391815185546875, 3.1736907958984375 ], "device": "cuda" }, "baseline_velocity_trials": [ { "velocity_delta": -36, "spectral_convergence": 0.8310962915420532, "log_band_mae": 0.4577361047267914, "waveform_rmse": 0.03823858126997948, "peak": 0.082489013671875 }, { "velocity_delta": -24, "spectral_convergence": 0.81913161277771, "log_band_mae": 0.45899340510368347, "waveform_rmse": 0.04240952432155609, "peak": 0.1371002197265625 }, { "velocity_delta": -12, "spectral_convergence": 0.9158480167388916, "log_band_mae": 0.49990421533584595, "waveform_rmse": 0.04917972907423973, "peak": 0.1988067626953125 }, { "velocity_delta": 0, "spectral_convergence": 1.1701204776763916, "log_band_mae": 0.5716279149055481, "waveform_rmse": 0.06068168580532074, "peak": 0.287872314453125 }, { "velocity_delta": 12, "spectral_convergence": 1.588895320892334, "log_band_mae": 0.6667157411575317, "waveform_rmse": 0.07582571357488632, "peak": 0.3860015869140625 } ], "velocity_trials": [ { "velocity_delta": -12, "spectral_convergence": 0.7651746869087219, "log_band_mae": 0.41147884726524353, "waveform_rmse": 0.041835229843854904, "peak": 0.16778564453125 }, { "velocity_delta": -6, "spectral_convergence": 0.7547203302383423, "log_band_mae": 0.40575388073921204, "waveform_rmse": 0.04339601844549179, "peak": 0.1864776611328125 }, { "velocity_delta": 6, "spectral_convergence": 0.7828538417816162, "log_band_mae": 0.41924867033958435, "waveform_rmse": 0.0466630682349205, "peak": 0.2800140380859375 }, { "velocity_delta": 12, "spectral_convergence": 0.8264999985694885, "log_band_mae": 0.4333062171936035, "waveform_rmse": 0.04959730803966522, "peak": 0.3137359619140625 } ], "wall_seconds": 458.2200195789337, "caveats": [ "These longer demo samples have no human intelligibility or ASR evaluation.", "Magnitude addition is approximate; phase interactions evaluated only on re-render.", "73 pitches, 3 velocities, 3 gate durations; short dictionary tails are truncated.", "Baseline is heuristic and has a different note count." ] }