Table 5. MATE training configurations on...

Table 5.

MATE training configurations on different data sets

Data set	AISHELL-1	TEDLIUM-2	LibriSpeech
Epochs	5	5	5
Batch size	4	4	4
Gradient accumulation	8	8	8
Learning rate	2e-5	2e-5	2e-5
Warm up Ratio	0.1	0.1	0.1

Sharing Unavailable