HyperAIHyperAI초신경
홈뉴스연구 논문튜토리얼데이터셋백과사전SOTALLM 모델GPU 랭킹컨퍼런스
전체 검색
소개
HyperAIHyperAI초신경

Speaker Identification On Voxceleb1

평가 지표

Accuracy
Top-1 (%)

평가 결과

이 벤치마크에서 각 모델의 성능 결과

모델 이름
Accuracy
Top-1 (%)
Paper TitleRepository
SSAST-PATCH64.264.2SSAST: Self-Supervised Audio Spectrogram Transformer
AutoSpeech (N=8,C=128)87.6687.66AutoSpeech: Neural Architecture Search for Speaker Recognition
SSAST-FRAME80.880.8SSAST: Self-Supervised Audio Spectrogram Transformer
ATST Base (ours)94.394.3ATST: Audio Representation Learning with Teacher-Student Transformer
M2D ratio=0.694.894.8Masked Modeling Duo: Learning Representations by Encouraging Both Networks to Model the Input
M2D/0.696.596.5Masked Modeling Duo: Towards a Universal Audio Pre-training Framework
AudioMAE (local)94.894.8Masked Autoencoders that Listen
MSM-MAE96.696.6Masked Modeling Duo: Towards a Universal Audio Pre-training Framework
AudioMAE (global)94.194.1Masked Autoencoders that Listen
M2D/0.796.396.3Masked Modeling Duo: Towards a Universal Audio Pre-training Framework
COLA37.737.7Contrastive Learning of General-Purpose Audio Representations
SSAMBA70.170.1SSAMBA: Self-Supervised Audio Representation Learning with Mamba State Space Model
0 of 12 row(s) selected.