Does conformer support batch_size > 1 in ASR task inference?

[Ht-zhang-xianyu]

I am trying use Conformer for Aishell inference in ASR task
the pretraiend or model config I use is the follow link:
https://github.com/espnet/espnet/tree/master/egs2/aishell/asr1#conformer--specaug--speed-perturbation-featsraw-n_fft512-hop_length128
But I found the API Speech2Text in bin/asr_inference.py only supports batch_size==1
Is there some demo showing how to inference the dataset with batch_size > 1?
Thanks

[sw005320]

Unfortunately, we don't support the case with batch_size > 1.