List your company

Utterance-level classification on ARCH (RAVDESS, EMOVO, AudioMNIST, SLURP) post-VQ 57

37.85RAVDESS AccuracyDTM-Codec@50Hz
23.768427.424231.0834.7358Jun 28, 2026

Evaluation results

MethodLinks
DTM-Codec@50HzSetting=VFR, |C|=16,3842026.0637.8536.0121.9414.5368.2668.057.110.9633.7929.89
DTM-Codec@40HzSetting=FFR, |C|=65,5362026.0637.535.2224.8320.0465.7365.6570.9533.7730.46
DTM-Codec@80HzSetting=VFR, |C|=16,3842026.0637.533.2523.8116.458.6258.136.970.8431.7327.16
X-Codec 2.0Setting=50Hz, |C|=65,536, Features=Wav2Vec2-BERT semantic features2026.0637.1532.8820.7515.5968.4968.157.741.2233.5329.46
DTM-Codec@80HzSetting=FFR, |C|=65,5362026.0636.8133.9326.1919.5670.8870.7670.9635.2231.3
BigCodecSetting=80Hz, |C|=8,1922026.0636.1134.4317.1812.3865.8465.747.671.0531.728.4
DTM-Codec@25HzSetting=FFR, |C|=65,5362026.0635.4233.5626.722.1365.0764.777.451.0633.6630.38
DTM-Codec@50HzSetting=FFR, |C|=65,5362026.0634.7231.6924.4918.7169.6369.557.231.0534.0230.25
DTM-Codec@40HzSetting=VFR, |C|=16,3842026.0632.9928.9721.0917.1270.6970.457.611.1533.0929.42
DTM-Codec@25HzSetting=VFR, |C|=16,3842026.0632.6428.6127.3822.7565.8465.457.61.1933.3729.5
VARSTokSetting=τ=0.8, |C|=4,0962026.0627.4324.0127.2121.3960.4260.187.621.1330.6726.68
WavTokenizerSetting=75Hz, |C|=4,0962026.0627.4322.520.4115.856.6256.187.120.8227.923.82
VARSTokSetting=τ=0.7, |C|=4,0962026.0627.0823.7425.6820.2761.3561.097.491.0730.426.54
WavTokenizerSetting=40Hz, |C|=4,0962026.0624.6521.826.1918.6950.12496.690.5326.9122.5
VARSTokSetting=τ=0.6, |C|=4,0962026.0624.3121.0324.8319.3462.3662.17.28129.6925.87