HyperAI超神経

ホームプラットフォームドキュメントニュース論文チュートリアルデータセット百科事典 SOTA LLMモデル GPU ランキング学会

サイトについて

日本語

HyperAI超神経

Speech Recognition On Slue

評価指標

VoxCeleb (Dev)

VoxCeleb (Test)

VoxPopuli (Dev)

VoxPopuli (Test)

評価結果

このベンチマークにおける各モデルのパフォーマンス結果

					Paper Title	Repository
W2V2-B-VP100K	29.9	33.4	21.6	22.4	SLUE: New Benchmark Tasks for Spoken Language Understanding Evaluation on Natural Speech
HuBERT-B-LS960	19.6	21.2	18.6	19.1	SLUE: New Benchmark Tasks for Spoken Language Understanding Evaluation on Natural Speech
W2V2-B-LS960	17.2	20.5	17.2	17.9	SLUE: New Benchmark Tasks for Spoken Language Understanding Evaluation on Natural Speech
W2V2-B-LS960 (+ in-domain LM)	15.2	18.2	14.6	15.2	SLUE: New Benchmark Tasks for Spoken Language Understanding Evaluation on Natural Speech
W2V2-B-LS960 (+ TED-LIUM 3 LM)	13.2	15.8	12.0	12.2	SLUE: New Benchmark Tasks for Spoken Language Understanding Evaluation on Natural Speech
W2V2-L-LL60K (+ in-domain LM)	11.8	13.8	12.0	12.5	SLUE: New Benchmark Tasks for Spoken Language Understanding Evaluation on Natural Speech
W2V2-L-LL60K	11.0	13.5	14.0	12.1	SLUE: New Benchmark Tasks for Spoken Language Understanding Evaluation on Natural Speech
W2V2-L-LL60K (+ TED-LIUM 3 LM)	9.1	10.8	9.1	9.3	SLUE: New Benchmark Tasks for Spoken Language Understanding Evaluation on Natural Speech

0 of 8 row(s) selected.