提交

kongdeqiang

2026-03-13 28ccfbfc51068a663a80764e14074df5edf2b5ba

 examples/aishell/transformer/run.sh

@@ -5,7 +5,7 @@

# general configuration
feats_dir="../DATA" #feature output dictionary
exp_dir="."
exp_dir=`pwd`
lang=zh
token_type=char
stage=0
@@ -15,9 +15,9 @@
nj=32

inference_device="cuda" #"cpu"
inference_checkpoint="model.pt"
inference_checkpoint="model.pt.avg10"
inference_scp="wav.scp"
inference_batch_size=32
inference_batch_size=1

# data
raw_data=../raw_data
@@ -26,6 +26,8 @@
# exp tag
tag="exp1"
workspace=`pwd`

master_port=12345

. utils/parse_options.sh || exit 1;

@@ -39,8 +41,9 @@
valid_set=dev
test_sets="dev test"

config=paraformer_conformer_12e_6d_2048_256.yaml
config=transformer_12e_6d_2048_256.yaml
model_dir="baseline_$(basename "${config}" .yaml)_${lang}_${token_type}_${tag}"



if [ ${stage} -le -1 ] && [ ${stop_stage} -ge -1 ]; then
@@ -77,7 +80,7 @@
    --config-name "${config}" \
    ++train_data_set_list="${feats_dir}/data/${train_set}/audio_datasets.jsonl" \
    ++cmvn_file="${feats_dir}/data/${train_set}/cmvn.json" \
    ++dataset_conf.num_workers=$nj

fi

token_list=${feats_dir}/data/${lang}_token_list/$token_type/tokens.txt
@@ -85,7 +88,7 @@
if [ ${stage} -le 2 ] && [ ${stop_stage} -ge 2 ]; then
    echo "stage 2: Dictionary Preparation"
    mkdir -p ${feats_dir}/data/${lang}_token_list/$token_type/
   

    echo "make a dictionary"
    echo "<blank>" > ${token_list}
    echo "<s>" >> ${token_list}
@@ -109,10 +112,12 @@
  log_file="${exp_dir}/exp/${model_dir}/train.log.txt.${current_time}"
  echo "log_file: ${log_file}"

  export CUDA_VISIBLE_DEVICES=$CUDA_VISIBLE_DEVICES
  gpu_num=$(echo $CUDA_VISIBLE_DEVICES | awk -F "," '{print NF}')
  torchrun \
  --nnodes 1 \
  --nproc_per_node ${gpu_num} \
  --master_port ${master_port} \
  ../../../funasr/bin/train.py \
  --config-path "${workspace}/conf" \
  --config-name "${config}" \
@@ -129,7 +134,7 @@
if [ ${stage} -le 5 ] && [ ${stop_stage} -ge 5 ]; then
  echo "stage 5: Inference"

  if ${inference_device} == "cuda"; then
  if [ ${inference_device} == "cuda" ]; then
      nj=$(echo $CUDA_VISIBLE_DEVICES | awk -F "," '{print NF}')
  else
      inference_batch_size=1
@@ -141,8 +146,9 @@

  for dset in ${test_sets}; do

    inference_dir="${exp_dir}/exp/${model_dir}/${inference_checkpoint}/${dset}"
    inference_dir="${exp_dir}/exp/${model_dir}/inference-${inference_checkpoint}/${dset}"
    _logdir="${inference_dir}/logdir"
    echo "inference_dir: ${inference_dir}"

    mkdir -p "${_logdir}"
    data_dir="${feats_dir}/data/${dset}"
@@ -154,7 +160,7 @@
    done
    utils/split_scp.pl "${key_file}" ${split_scps}

    gpuid_list_array=(${gpuid_list//,/ })
    gpuid_list_array=(${CUDA_VISIBLE_DEVICES//,/ })
    for JOB in $(seq ${nj}); do
        {
          id=$((JOB-1))
@@ -170,7 +176,9 @@
          ++input="${_logdir}/keys.${JOB}.scp" \
          ++output_dir="${inference_dir}/${JOB}" \
          ++device="${inference_device}" \
          ++batch_size="${inference_batch_size}"
          ++ncpu=1 \
          ++disable_log=true \
          ++batch_size="${inference_batch_size}" &> ${_logdir}/log.${JOB}.txt
        }&

    done
@@ -186,10 +194,10 @@
    done

    echo "Computing WER ..."
    cp ${inference_dir}/1best_recog/text ${inference_dir}/1best_recog/text.proc
    cp ${data_dir}/text ${inference_dir}/1best_recog/text.ref
    python utils/postprocess_text_zh.py ${inference_dir}/1best_recog/text ${inference_dir}/1best_recog/text.proc
    python utils/postprocess_text_zh.py  ${data_dir}/text ${inference_dir}/1best_recog/text.ref
    python utils/compute_wer.py ${inference_dir}/1best_recog/text.ref ${inference_dir}/1best_recog/text.proc ${inference_dir}/1best_recog/text.cer
    tail -n 3 ${inference_dir}/1best_recog/text.cer
  done

fi
fi