mirror of
https://github.com/k2-fsa/icefall.git
synced 2025-09-10 09:34:39 +00:00
updated
This commit is contained in:
parent
a704a2758b
commit
106007600e
@ -256,8 +256,6 @@ if [ $stage -le 11 ] && [ $stop_stage -ge 11 ]; then
|
|||||||
log "Abort! Please run ../../wenetspeech/ASR/prepare.sh"
|
log "Abort! Please run ../../wenetspeech/ASR/prepare.sh"
|
||||||
exit 1
|
exit 1
|
||||||
fi
|
fi
|
||||||
|
|
||||||
|
|
||||||
fi
|
fi
|
||||||
|
|
||||||
log "Dataset: KeSpeech"
|
log "Dataset: KeSpeech"
|
||||||
@ -310,3 +308,20 @@ if [ $stage -le 12 ] && [ $stop_stage -ge 12 ]; then
|
|||||||
touch data/fbank/.kespeech.done
|
touch data/fbank/.kespeech.done
|
||||||
fi
|
fi
|
||||||
fi
|
fi
|
||||||
|
|
||||||
|
if [ $stage -le 13 ] && [ $stop_stage -ge 13 ]; then
|
||||||
|
log "Stage 13: BPE model training"
|
||||||
|
./local/prepare_for_bpe_model.py --lang-dir ./data/lang_char --text ./data/lang_char/text
|
||||||
|
|
||||||
|
for vocab_size in ${vocab_sizes[@]}; do
|
||||||
|
lang_dir=data/lang_bpe_${vocab_size}
|
||||||
|
|
||||||
|
mkdir -p $lang_dir
|
||||||
|
./local/train_bpe_model.py \
|
||||||
|
--lang-dir $lang_dir \
|
||||||
|
--transcript ./data/lang_char/transcript_chars.txt \
|
||||||
|
--vocab-size $vocab_size
|
||||||
|
done
|
||||||
|
|
||||||
|
./local/train_bpe_model.py --lang-dir ./data/lang_bpe_${vocab_size}
|
||||||
|
fi
|
Loading…
x
Reference in New Issue
Block a user