From 33d3d2084403fd34b79c835d2f2fe04f6cd8f738 Mon Sep 17 00:00:00 2001
From: 游雁 <zhifu.gzf@alibaba-inc.com>
Date: 星期三, 13 九月 2023 09:33:54 +0800
Subject: [PATCH] Merge branch 'main' of github.com:alibaba-damo-academy/FunASR add
---
egs/aishell/paraformer/run.sh | 68 ++++++++++++++++++++++------------
1 files changed, 44 insertions(+), 24 deletions(-)
diff --git a/egs/aishell/paraformer/run.sh b/egs/aishell/paraformer/run.sh
index bcfda14..7d79211 100755
--- a/egs/aishell/paraformer/run.sh
+++ b/egs/aishell/paraformer/run.sh
@@ -16,26 +16,23 @@
feats_dir="../DATA" #feature output dictionary
exp_dir="."
lang=zh
-dumpdir=dump/fbank
-feats_type=fbank
token_type=char
-scp=wav.scp
type=sound
-stage=1
-stop_stage=1
+scp=wav.scp
+speed_perturb="0.9 1.0 1.1"
+stage=0
+stop_stage=5
# feature configuration
feats_dim=80
-sample_frequency=16000
nj=64
-speed_perturb="0.9,1.0,1.1"
# data
-raw_data=
+raw_data=../raw_data
data_url=www.openslr.org/resources/33
# exp tag
-tag=""
+tag="exp1"
. utils/parse_options.sh || exit 1;
@@ -50,7 +47,7 @@
test_sets="dev test"
asr_config=conf/train_asr_paraformer_conformer_12e_6d_2048_256.yaml
-model_dir="baseline_$(basename "${asr_config}" .yaml)_${feats_type}_${lang}_${token_type}_${tag}"
+model_dir="baseline_$(basename "${asr_config}" .yaml)_${lang}_${token_type}_${tag}"
inference_config=conf/decode_asr_transformer_noctc_1best.yaml
inference_asr_model=valid.acc.ave_10best.pb
@@ -86,33 +83,36 @@
done
fi
-feat_train_dir=${feats_dir}/${dumpdir}/train; mkdir -p ${feat_train_dir}
if [ ${stage} -le 1 ] && [ ${stop_stage} -ge 1 ]; then
echo "stage 1: Feature and CMVN Generation"
- utils/compute_cmvn.sh --cmd "$train_cmd" --nj $nj --feats_dim ${feats_dim} ${feats_dir}/data/${train_set}
+ utils/compute_cmvn.sh --fbankdir ${feats_dir}/data/${train_set} --cmd "$train_cmd" --nj $nj --feats_dim ${feats_dim} --config_file "$asr_config" --scale 1.0
fi
-token_list=${feats_dir}/data/${lang}_token_list/char/tokens.txt
+token_list=${feats_dir}/data/${lang}_token_list/$token_type/tokens.txt
echo "dictionary: ${token_list}"
if [ ${stage} -le 2 ] && [ ${stop_stage} -ge 2 ]; then
echo "stage 2: Dictionary Preparation"
- mkdir -p ${feats_dir}/data/${lang}_token_list/char/
+ mkdir -p ${feats_dir}/data/${lang}_token_list/$token_type/
echo "make a dictionary"
echo "<blank>" > ${token_list}
echo "<s>" >> ${token_list}
echo "</s>" >> ${token_list}
- utils/text2token.py -s 1 -n 1 --space "" ${feats_dir}/data/train/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}' >> ${token_list}
- num_token=$(cat ${token_list} | wc -l)
echo "<unk>" >> ${token_list}
- vocab_size=$(cat ${token_list} | wc -l)
fi
-# Training Stage
+# LM Training Stage
world_size=$gpu_num # run on one machine
if [ ${stage} -le 3 ] && [ ${stop_stage} -ge 3 ]; then
- echo "stage 3: Training"
+ echo "stage 3: LM Training"
+fi
+
+# ASR Training Stage
+world_size=$gpu_num # run on one machine
+if [ ${stage} -le 4 ] && [ ${stop_stage} -ge 4 ]; then
+ echo "stage 4: ASR Training"
mkdir -p ${exp_dir}/exp/${model_dir}
mkdir -p ${exp_dir}/exp/${model_dir}/log
INIT_FILE=${exp_dir}/exp/${model_dir}/ddp_init
@@ -130,12 +130,15 @@
--task_name asr \
--gpu_id $gpu_id \
--use_preprocessor true \
- --token_type char \
+ --token_type $token_type \
--token_list $token_list \
+ --dataset_type small \
--data_dir ${feats_dir}/data \
--train_set ${train_set} \
--valid_set ${valid_set} \
- --cmvn_file ${feats_dir}/cmvn/cmvn.mvn \
+ --data_file_names "wav.scp,text" \
+ --cmvn_file ${feats_dir}/data/${train_set}/cmvn/am.mvn \
+ --speed_perturb ${speed_perturb} \
--resume true \
--output_dir ${exp_dir}/exp/${model_dir} \
--config $asr_config \
@@ -151,8 +154,8 @@
fi
# Testing Stage
-if [ ${stage} -le 4 ] && [ ${stop_stage} -ge 4 ]; then
- echo "stage 4: Inference"
+if [ ${stage} -le 5 ] && [ ${stop_stage} -ge 5 ]; then
+ echo "stage 5: Inference"
for dset in ${test_sets}; do
asr_exp=${exp_dir}/exp/${model_dir}
inference_tag="$(basename "${inference_config}" .yaml)"
@@ -163,7 +166,7 @@
exit 0
fi
mkdir -p "${_logdir}"
- _data="${feats_dir}/${dumpdir}/${dset}"
+ _data="${feats_dir}/data/${dset}"
key_file=${_data}/${scp}
num_scp_file="$(<${key_file} wc -l)"
_nj=$([ $inference_nj -le $num_scp_file ] && echo "$inference_nj" || echo "$num_scp_file")
@@ -184,6 +187,7 @@
--njob ${njob} \
--gpuid_list ${gpuid_list} \
--data_path_and_name_and_type "${_data}/${scp},speech,${type}" \
+ --cmvn_file ${feats_dir}/data/${train_set}/cmvn/am.mvn \
--key_file "${_logdir}"/keys.JOB.scp \
--asr_train_config "${asr_exp}"/config.yaml \
--asr_model_file "${asr_exp}"/"${inference_asr_model}" \
@@ -204,4 +208,20 @@
tail -n 3 ${_dir}/text.cer > ${_dir}/text.cer.txt
cat ${_dir}/text.cer.txt
done
+fi
+
+# Prepare files for ModelScope fine-tuning and inference
+if [ ${stage} -le 6 ] && [ ${stop_stage} -ge 6 ]; then
+ echo "stage 6: ModelScope Preparation"
+ cp ${feats_dir}/data/${train_set}/cmvn/am.mvn ${exp_dir}/exp/${model_dir}/am.mvn
+ vocab_size=$(cat ${token_list} | wc -l)
+ python utils/gen_modelscope_configuration.py \
+ --am_model_name $inference_asr_model \
+ --mode paraformer \
+ --model_name paraformer \
+ --dataset aishell \
+ --output_dir $exp_dir/exp/$model_dir \
+ --vocab_size $vocab_size \
+ --nat _nat \
+ --tag $tag
fi
\ No newline at end of file
--
Gitblit v1.9.1