|
W2V2-L-LL60K (pipeline approach, uses LM)
|
SLUE: New Benchmark Tasks for Spoken Language Und…
|
69.60
|
2021-11-19
|
|
|
W2V2-B-LS960 (pipeline approach, uses LM)
|
SLUE: New Benchmark Tasks for Spoken Language Und…
|
68.00
|
2021-11-19
|
|
|
Wav2Seq (from HuBERT-large)
|
Wav2Seq: Pre-training Speech-to-Text Encoder-Deco…
|
65.40
|
2022-05-02
|
|
|
W2V2-L-LL60K (e2e approach, uses LM)
|
SLUE: New Benchmark Tasks for Spoken Language Und…
|
64.80
|
2021-11-19
|
|
|
W2V2-B-LS960 (e2e approach, uses LM)
|
SLUE: New Benchmark Tasks for Spoken Language Und…
|
63.40
|
2021-11-19
|
|
|
HuBERT-B-LS960 (e2e approach, uses LM)
|
SLUE: New Benchmark Tasks for Spoken Language Und…
|
61.90
|
2021-11-19
|
|
|
W2V2-B-VP100K (e2e approach, uses LM)
|
SLUE: New Benchmark Tasks for Spoken Language Und…
|
61.80
|
2021-11-19
|
|
|
W2V2-L-LL60K (pipeline approach)
|
SLUE: New Benchmark Tasks for Spoken Language Und…
|
57.80
|
2021-11-19
|
|
|
W2V2-L-LL60K (e2e approach)
|
SLUE: New Benchmark Tasks for Spoken Language Und…
|
50.90
|
2021-11-19
|
|
|
W2V2-B-LS960 (e2e approach)
|
SLUE: New Benchmark Tasks for Spoken Language Und…
|
50.20
|
2021-11-19
|
|
|
HuBERT-B-LS960 (e2e approach)
|
SLUE: New Benchmark Tasks for Spoken Language Und…
|
49.80
|
2021-11-19
|
|
|
W2V2-B-LS960 (pipeline approach)
|
SLUE: New Benchmark Tasks for Spoken Language Und…
|
49.50
|
2021-11-19
|
|
|
W2V2-B-VP100K (e2e approach)
|
SLUE: New Benchmark Tasks for Spoken Language Und…
|
47.90
|
2021-11-19
|
|