speech_recognition/FunASR
1
0
Fork 0
mirror of https://github.com/modelscope/FunASR synced 2025-09-15 14:48:36 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

update repo

Browse Source
This commit is contained in:
嘉渊 2023-05-15 14:54:28 +08:00
parent 8a5c12ef29
commit bf704cc031
1 changed files with 10 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

17
egs/aishell2/conformer/run.sh
View File

@ -20,6 +20,7 @@ token_type=char
type=sound type=sound
scp=wav.scp scp=wav.scp
speed_perturb="0.9 1.0 1.1" speed_perturb="0.9 1.0 1.1"
dataset_type=large
stage=2 stage=2
stop_stage=2 stop_stage=2
@ -99,7 +100,7 @@ if [ ${stage} -le 2 ] && [ ${stop_stage} -ge 2 ]; then
echo "<blank>" > ${token_list} echo "<blank>" > ${token_list}
echo "<s>" >> ${token_list} echo "<s>" >> ${token_list}
echo "</s>" >> ${token_list} echo "</s>" >> ${token_list}
tools/text2token.py -s 1 -n 1 --space "" ${feats_dir}/data/${train_set}/text | cut -f 2- -d" " | tr " " "\n" \ utils/text2token.py -s 1 -n 1 --space "" ${feats_dir}/data/${train_set}/text | cut -f 2- -d" " | tr " " "\n" \
| sort | uniq | grep -a -v -e '^\s*$' | awk '{print $0}' >> ${token_list} | sort | uniq | grep -a -v -e '^\s*$' | awk '{print $0}' >> ${token_list}
echo "<unk>" >> ${token_list} echo "<unk>" >> ${token_list}
mkdir -p ${feats_dir}/asr_stats_fbank_zh_char/${train_set} mkdir -p ${feats_dir}/asr_stats_fbank_zh_char/${train_set}
@ -123,21 +124,23 @@ if [ ${stage} -le 3 ] && [ ${stop_stage} -ge 3 ]; then
rank=$i rank=$i
local_rank=$i local_rank=$i
gpu_id=$(echo $CUDA_VISIBLE_DEVICES | cut -d',' -f$[$i+1]) gpu_id=$(echo $CUDA_VISIBLE_DEVICES | cut -d',' -f$[$i+1])
asr_train.py \ train.py \
--task_name asr \
--gpu_id $gpu_id \ --gpu_id $gpu_id \
--use_preprocessor true \ --use_preprocessor true \
--dataset_type $dataset_type \
--token_type char \ --token_type char \
--token_list $token_list \ --token_list $token_list \
--train_data_file $feats_dir/$dumpdir/${train_set}/data.list \ --data_dir ${feats_dir}/data \
--valid_data_file $feats_dir/$dumpdir/${valid_set}/data.list \ --train_set ${train_set} \
--valid_set ${valid_set} \
--cmvn_file ${feats_dir}/data/${train_set}/cmvn/cmvn.mvn \
--speed_perturb ${speed_perturb} \
--dataset_type $dataset_type \
--resume true \ --resume true \
--output_dir ${exp_dir}/exp/${model_dir} \ --output_dir ${exp_dir}/exp/${model_dir} \
--config $asr_config \ --config $asr_config \
--input_size $feats_dim \
--ngpu $gpu_num \ --ngpu $gpu_num \
--num_worker_count $count \ --num_worker_count $count \
--multiprocessing_distributed true \
--dist_init_method $init_method \ --dist_init_method $init_method \
--dist_world_size $world_size \ --dist_world_size $world_size \
--dist_rank $rank \ --dist_rank $rank \