Alexei Baevski

thanks for investigating @zhxchen17 . Shall we open another issue for the pytorch test breakage and reference it here?

you probably need to change the striding in feature extractor so that your representation encode approximately the same amount of audio as it would if they were 16khz. i am...

you can load the checkpoint now but it appears the checkpoint has the wrong task (probably created before audio_finetuning was split off from audio_pretraining), and the remedy there is to...

it works for me. are you sure you have the latest fairseq version installed? try updating to master and then running something like this from the checked out dir: ```shell...

did you try following the instruction on the last line?

i think kenlm by default only supports up to 5gram, and then you need to recompile it with diff flags to support more grams than that. dunno if that is...