{ "input": "examples/zh/source.wav", "seconds": 7.6, "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": 0.8310245275497437, "log_band_mae": 0.4479677379131317, "waveform_rmse": 0.037389934062957764, "peak": 0.0824432373046875, "notes": 445 }, "dictionary": { "spectral_convergence": 0.7704614400863647, "log_band_mae": 0.4111631214618683, "waveform_rmse": 0.03957610949873924, "peak": 0.17767333984375, "notes": 123 }, "refined": { "spectral_convergence": 0.7986038327217102, "log_band_mae": 0.3923003077507019, "waveform_rmse": 0.03847990930080414, "peak": 0.20697021484375, "notes": 117 } }, "render_repeatability": { "spectral_convergence": 0.00047312359674833715, "log_band_loss_delta": 3.2186508178710938e-06 }, "pursuit": { "accepted": 123, "surrogate_improvements": [ 5873.54931640625, 4905.5556640625, 4766.67578125, 3959.75, 3534.9482421875, 3442.66455078125, 3219.97216796875, 3065.7041015625, 1935.3251953125, 1933.53662109375, 1805.196044921875, 1793.760986328125, 1728.0030517578125, 1548.9722900390625, 1492.723876953125, 1421.999755859375, 1394.755615234375, 1346.2169189453125, 1329.601318359375, 1275.04443359375, 1201.2552490234375, 1187.4041748046875, 1177.69775390625, 1154.324951171875, 1132.4447021484375, 971.7203369140625, 953.567138671875, 911.7687377929688, 864.153076171875, 861.8289794921875, 842.8236083984375, 723.0029296875, 690.95458984375, 670.2884521484375, 647.384033203125, 643.4237670898438, 596.17041015625, 591.5808715820312, 590.9668579101562, 576.955810546875, 555.5059204101562, 516.6516723632812, 514.6587524414062, 514.279052734375, 504.91949462890625, 411.443115234375, 400.2333984375, 390.39208984375, 380.6637878417969, 370.4420166015625, 363.265380859375, 346.45257568359375, 342.1094970703125, 331.36553955078125, 325.9708251953125, 324.1048278808594, 323.36859130859375, 314.0655517578125, 312.26904296875, 305.7211608886719, 290.40826416015625, 283.8734130859375, 276.9443359375, 269.7228698730469, 242.9066162109375, 240.60430908203125, 240.25360107421875, 229.510986328125, 228.459716796875, 221.15936279296875, 187.22242736816406, 183.34634399414062, 182.0286865234375, 181.97256469726562, 175.39633178710938, 141.53753662109375, 140.17288208007812, 138.97979736328125, 137.34613037109375, 135.32421875, 132.63221740722656, 130.34976196289062, 123.92070007324219, 122.67015075683594, 117.83511352539062, 115.94491577148438, 112.36289978027344, 112.08888244628906, 111.0322265625, 106.10653686523438, 102.9644775390625, 102.46817016601562, 97.46054077148438, 92.73184204101562, 92.45913696289062, 91.44773864746094, 81.18904113769531, 76.4394302368164, 72.32395935058594, 68.6805191040039, 67.99607849121094, 61.383392333984375, 57.82987594604492, 51.87420654296875, 51.49356460571289, 41.40382385253906, 40.713409423828125, 39.9324951171875, 38.902130126953125, 32.54479217529297, 32.269126892089844, 32.22667694091797, 28.96026611328125, 28.6995849609375, 28.284168243408203, 26.704544067382812, 23.561458587646484, 19.500450134277344, 17.845294952392578, 16.587181091308594, 12.098899841308594, 11.388519287109375, 11.218666076660156 ], "device": "cuda" }, "baseline_velocity_trials": [ { "velocity_delta": -36, "spectral_convergence": 0.831043004989624, "log_band_mae": 0.44797632098197937, "waveform_rmse": 0.03738969564437866, "peak": 0.082427978515625 }, { "velocity_delta": -24, "spectral_convergence": 0.8230495452880859, "log_band_mae": 0.45736628770828247, "waveform_rmse": 0.04087916389107704, "peak": 0.129547119140625 }, { "velocity_delta": -12, "spectral_convergence": 0.9122947454452515, "log_band_mae": 0.5070769786834717, "waveform_rmse": 0.04709188640117645, "peak": 0.189208984375 }, { "velocity_delta": 0, "spectral_convergence": 1.1454532146453857, "log_band_mae": 0.5840789675712585, "waveform_rmse": 0.05760560557246208, "peak": 0.281158447265625 }, { "velocity_delta": 12, "spectral_convergence": 1.5337176322937012, "log_band_mae": 0.6722043752670288, "waveform_rmse": 0.07145288586616516, "peak": 0.366973876953125 } ], "velocity_trials": [ { "velocity_delta": -12, "spectral_convergence": 0.7999444603919983, "log_band_mae": 0.41019579768180847, "waveform_rmse": 0.03728783130645752, "peak": 0.13568115234375 }, { "velocity_delta": -6, "spectral_convergence": 0.7826135754585266, "log_band_mae": 0.4063299596309662, "waveform_rmse": 0.038131602108478546, "peak": 0.151519775390625 }, { "velocity_delta": 6, "spectral_convergence": 0.7779667377471924, "log_band_mae": 0.4148591458797455, "waveform_rmse": 0.042012300342321396, "peak": 0.1972198486328125 }, { "velocity_delta": 12, "spectral_convergence": 0.7948122024536133, "log_band_mae": 0.4231906533241272, "waveform_rmse": 0.04392588511109352, "peak": 0.2220611572265625 } ], "wall_seconds": 250.1855344772339, "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." ] }