diff --git a/paddleaudio/setup.py b/paddleaudio/setup.py
index 6c757d336..930f86e41 100644
--- a/paddleaudio/setup.py
+++ b/paddleaudio/setup.py
@@ -61,6 +61,7 @@ def remove_version_py(filename='paddleaudio/__init__.py'):
             if "__version__" not in line:
                 f.write(line)
 
+
 remove_version_py()
 write_version_py()
 
diff --git a/paddlespeech/cli/utils.py b/paddlespeech/cli/utils.py
index d7dcc90c7..f7d64b9a9 100644
--- a/paddlespeech/cli/utils.py
+++ b/paddlespeech/cli/utils.py
@@ -192,7 +192,7 @@ class ConfigCache:
             try:
                 cfg = yaml.load(file, Loader=yaml.FullLoader)
                 self._data.update(cfg)
-            except:
+            except Exception as e:
                 self.flush()
 
     @property
diff --git a/paddlespeech/server/bin/paddlespeech_server.py b/paddlespeech/server/bin/paddlespeech_server.py
index 7e7f03b2d..f6a7f4295 100644
--- a/paddlespeech/server/bin/paddlespeech_server.py
+++ b/paddlespeech/server/bin/paddlespeech_server.py
@@ -174,7 +174,7 @@ class ServerStatsExecutor():
                     "Failed to get the table of TTS pretrained models supported in the service."
                 )
                 return False
-        
+
         elif self.task == 'cls':
             try:
                 from paddlespeech.cli.cls.infer import pretrained_models
diff --git a/speechx/examples/README.md b/speechx/examples/README.md
index 28001ad9c..941c4272d 100644
--- a/speechx/examples/README.md
+++ b/speechx/examples/README.md
@@ -13,4 +13,4 @@ Example to play `decoder`:
 ```
 pushd decoder
 bash run.sh
-```
\ No newline at end of file
+```
diff --git a/speechx/examples/feat/linear_spectrogram_main.cc b/speechx/examples/feat/linear_spectrogram_main.cc
index f137a52cc..9ed4d6f93 100644
--- a/speechx/examples/feat/linear_spectrogram_main.cc
+++ b/speechx/examples/feat/linear_spectrogram_main.cc
@@ -164,8 +164,8 @@ int main(int argc, char* argv[]) {
     // test feature linear_spectorgram: wave --> decibel_normalizer --> hanning
     // window -->linear_spectrogram --> cmvn
     int32 num_done = 0, num_err = 0;
-    //std::unique_ptr<ppspeech::FeatureExtractorInterface> data_source(new
-     //ppspeech::RawDataCache());
+    // std::unique_ptr<ppspeech::FeatureExtractorInterface> data_source(new
+    // ppspeech::RawDataCache());
     std::unique_ptr<ppspeech::FeatureExtractorInterface> data_source(
         new ppspeech::RawAudioCache());
 
diff --git a/speechx/speechx/decoder/ctc_beam_search_decoder.cc b/speechx/speechx/decoder/ctc_beam_search_decoder.cc
index 582f2c950..84f1453c0 100644
--- a/speechx/speechx/decoder/ctc_beam_search_decoder.cc
+++ b/speechx/speechx/decoder/ctc_beam_search_decoder.cc
@@ -52,14 +52,14 @@ CTCBeamSearch::CTCBeamSearch(const CTCBeamSearchOptions& opts)
 }
 
 void CTCBeamSearch::Reset() {
-    //num_frame_decoded_ = 0;
-    //ResetPrefixes();
+    // num_frame_decoded_ = 0;
+    // ResetPrefixes();
     InitDecoder();
 }
 
 void CTCBeamSearch::InitDecoder() {
     num_frame_decoded_ = 0;
-    //ResetPrefixes();
+    // ResetPrefixes();
     prefixes_.clear();
 
     root_ = std::make_shared<PathTrie>();
diff --git a/speechx/speechx/frontend/feature_cache.cc b/speechx/speechx/frontend/feature_cache.cc
index b353df164..d23b3a8b2 100644
--- a/speechx/speechx/frontend/feature_cache.cc
+++ b/speechx/speechx/frontend/feature_cache.cc
@@ -29,8 +29,7 @@ FeatureCache::FeatureCache(
     base_extractor_ = std::move(base_extractor);
 }
 
-void FeatureCache::Accept(
-    const kaldi::VectorBase<kaldi::BaseFloat>& inputs) {
+void FeatureCache::Accept(const kaldi::VectorBase<kaldi::BaseFloat>& inputs) {
     base_extractor_->Accept(inputs);
     // feed current data
     bool result = false;
diff --git a/speechx/speechx/frontend/feature_cache.h b/speechx/speechx/frontend/feature_cache.h
index 459134ee8..e52d8b298 100644
--- a/speechx/speechx/frontend/feature_cache.h
+++ b/speechx/speechx/frontend/feature_cache.h
@@ -24,8 +24,7 @@ class FeatureCache : public FeatureExtractorInterface {
     explicit FeatureCache(
         int32 max_size = kint16max,
         std::unique_ptr<FeatureExtractorInterface> base_extractor = NULL);
-    virtual void Accept(
-        const kaldi::VectorBase<kaldi::BaseFloat>& inputs);
+    virtual void Accept(const kaldi::VectorBase<kaldi::BaseFloat>& inputs);
     // feats dim = num_frames * feature_dim
     virtual bool Read(kaldi::Vector<kaldi::BaseFloat>* feats);
     // feature cache only cache feature which from base extractor
diff --git a/speechx/speechx/frontend/feature_extractor_controller.h b/speechx/speechx/frontend/feature_extractor_controller.h
index 5860f391c..0544a1e29 100644
--- a/speechx/speechx/frontend/feature_extractor_controller.h
+++ b/speechx/speechx/frontend/feature_extractor_controller.h
@@ -11,4 +11,3 @@
 // WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 // See the License for the specific language governing permissions and
 // limitations under the License.
-
diff --git a/speechx/speechx/frontend/feature_extractor_controller_impl.h b/speechx/speechx/frontend/feature_extractor_controller_impl.h
index 5860f391c..0544a1e29 100644
--- a/speechx/speechx/frontend/feature_extractor_controller_impl.h
+++ b/speechx/speechx/frontend/feature_extractor_controller_impl.h
@@ -11,4 +11,3 @@
 // WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 // See the License for the specific language governing permissions and
 // limitations under the License.
-
diff --git a/speechx/speechx/frontend/feature_extractor_interface.h b/speechx/speechx/frontend/feature_extractor_interface.h
index cb6fec1b8..3668fbda7 100644
--- a/speechx/speechx/frontend/feature_extractor_interface.h
+++ b/speechx/speechx/frontend/feature_extractor_interface.h
@@ -21,10 +21,9 @@ namespace ppspeech {
 
 class FeatureExtractorInterface {
   public:
-    // accept input data, accept feature or raw waves which decided 
+    // accept input data, accept feature or raw waves which decided
     // by the base_extractor
-    virtual void Accept(
-        const kaldi::VectorBase<kaldi::BaseFloat>& inputs) = 0;
+    virtual void Accept(const kaldi::VectorBase<kaldi::BaseFloat>& inputs) = 0;
     // get the processed result
     // the length of output = feature_row * feature_dim,
     // the Matrix is squashed into Vector
diff --git a/speechx/speechx/frontend/linear_spectrogram.cc b/speechx/speechx/frontend/linear_spectrogram.cc
index b8a18e028..41bc8743a 100644
--- a/speechx/speechx/frontend/linear_spectrogram.cc
+++ b/speechx/speechx/frontend/linear_spectrogram.cc
@@ -57,8 +57,9 @@ bool LinearSpectrogram::Read(Vector<BaseFloat>* feats) {
     if (flag == false || input_feats.Dim() == 0) return false;
 
     vector<BaseFloat> input_feats_vec(input_feats.Dim());
-    std::memcpy(input_feats_vec.data(), input_feats.Data(), 
-        input_feats.Dim()*sizeof(BaseFloat));
+    std::memcpy(input_feats_vec.data(),
+                input_feats.Data(),
+                input_feats.Dim() * sizeof(BaseFloat));
     vector<vector<BaseFloat>> result;
     Compute(input_feats_vec, result);
     int32 feat_size = 0;
@@ -86,10 +87,10 @@ bool LinearSpectrogram::NumpyFft(vector<BaseFloat>* v,
                                  vector<BaseFloat>* img) const {
     Vector<BaseFloat> v_tmp;
     v_tmp.Resize(v->size());
-    std::memcpy(v_tmp.Data(), v->data(), sizeof(BaseFloat)*(v->size()));
+    std::memcpy(v_tmp.Data(), v->data(), sizeof(BaseFloat) * (v->size()));
     RealFft(&v_tmp, true);
     v->resize(v_tmp.Dim());
-    std::memcpy(v->data(), v_tmp.Data(), sizeof(BaseFloat)*(v->size()));
+    std::memcpy(v->data(), v_tmp.Data(), sizeof(BaseFloat) * (v->size()));
 
     real->push_back(v->at(0));
     img->push_back(0);
diff --git a/speechx/speechx/frontend/linear_spectrogram.h b/speechx/speechx/frontend/linear_spectrogram.h
index b2bb414d5..ffdfbbe92 100644
--- a/speechx/speechx/frontend/linear_spectrogram.h
+++ b/speechx/speechx/frontend/linear_spectrogram.h
@@ -38,16 +38,13 @@ class LinearSpectrogram : public FeatureExtractorInterface {
     explicit LinearSpectrogram(
         const LinearSpectrogramOptions& opts,
         std::unique_ptr<FeatureExtractorInterface> base_extractor);
-    virtual void Accept(
-        const kaldi::VectorBase<kaldi::BaseFloat>& inputs);
+    virtual void Accept(const kaldi::VectorBase<kaldi::BaseFloat>& inputs);
     virtual bool Read(kaldi::Vector<kaldi::BaseFloat>* feats);
     // the dim_ is the dim of single frame feature
     virtual size_t Dim() const { return dim_; }
     virtual void SetFinished() { base_extractor_->SetFinished(); }
     virtual bool IsFinished() const { return base_extractor_->IsFinished(); }
-    virtual void Reset() {
-        base_extractor_->Reset();
-    }
+    virtual void Reset() { base_extractor_->Reset(); }
 
   private:
     void Hanning(std::vector<kaldi::BaseFloat>* data) const;
diff --git a/speechx/speechx/frontend/normalizer.cc b/speechx/speechx/frontend/normalizer.cc
index 285c8e03e..1adddb401 100644
--- a/speechx/speechx/frontend/normalizer.cc
+++ b/speechx/speechx/frontend/normalizer.cc
@@ -34,14 +34,12 @@ DecibelNormalizer::DecibelNormalizer(
     dim_ = 1;
 }
 
-void DecibelNormalizer::Accept(
-    const kaldi::VectorBase<BaseFloat>& waves) {
+void DecibelNormalizer::Accept(const kaldi::VectorBase<BaseFloat>& waves) {
     base_extractor_->Accept(waves);
 }
 
 bool DecibelNormalizer::Read(kaldi::Vector<BaseFloat>* waves) {
-    if (base_extractor_->Read(waves) == false || 
-        waves->Dim() == 0) {
+    if (base_extractor_->Read(waves) == false || waves->Dim() == 0) {
         return false;
     }
     Compute(waves);
@@ -88,7 +86,8 @@ bool DecibelNormalizer::Compute(VectorBase<BaseFloat>* waves) const {
         item *= std::pow(10.0, gain / 20.0);
     }
 
-    std::memcpy(waves->Data(), samples.data(), sizeof(BaseFloat)*samples.size());
+    std::memcpy(
+        waves->Data(), samples.data(), sizeof(BaseFloat) * samples.size());
     return true;
 }
 
diff --git a/speechx/speechx/frontend/normalizer.h b/speechx/speechx/frontend/normalizer.h
index 24542ebad..352d1e167 100644
--- a/speechx/speechx/frontend/normalizer.h
+++ b/speechx/speechx/frontend/normalizer.h
@@ -45,16 +45,13 @@ class DecibelNormalizer : public FeatureExtractorInterface {
     explicit DecibelNormalizer(
         const DecibelNormalizerOptions& opts,
         std::unique_ptr<FeatureExtractorInterface> base_extractor);
-    virtual void Accept(
-        const kaldi::VectorBase<kaldi::BaseFloat>& waves);
+    virtual void Accept(const kaldi::VectorBase<kaldi::BaseFloat>& waves);
     virtual bool Read(kaldi::Vector<kaldi::BaseFloat>* waves);
     // noramlize audio, the dim is 1.
     virtual size_t Dim() const { return dim_; }
     virtual void SetFinished() { base_extractor_->SetFinished(); }
     virtual bool IsFinished() const { return base_extractor_->IsFinished(); }
-    virtual void Reset() {
-        base_extractor_->Reset();
-    }
+    virtual void Reset() { base_extractor_->Reset(); }
 
   private:
     bool Compute(kaldi::VectorBase<kaldi::BaseFloat>* waves) const;
@@ -69,8 +66,7 @@ class CMVN : public FeatureExtractorInterface {
   public:
     explicit CMVN(std::string cmvn_file,
                   std::unique_ptr<FeatureExtractorInterface> base_extractor);
-    virtual void Accept(
-        const kaldi::VectorBase<kaldi::BaseFloat>& inputs);
+    virtual void Accept(const kaldi::VectorBase<kaldi::BaseFloat>& inputs);
 
     // the length of feats = feature_row * feature_dim,
     // the Matrix is squashed into Vector
@@ -79,9 +75,7 @@ class CMVN : public FeatureExtractorInterface {
     virtual size_t Dim() const { return dim_; }
     virtual void SetFinished() { base_extractor_->SetFinished(); }
     virtual bool IsFinished() const { return base_extractor_->IsFinished(); }
-    virtual void Reset() {
-        base_extractor_->Reset();
-    }
+    virtual void Reset() { base_extractor_->Reset(); }
 
   private:
     void Compute(kaldi::VectorBase<kaldi::BaseFloat>* feats) const;
diff --git a/speechx/speechx/frontend/raw_audio.cc b/speechx/speechx/frontend/raw_audio.cc
index 0f3d83ec0..21f643628 100644
--- a/speechx/speechx/frontend/raw_audio.cc
+++ b/speechx/speechx/frontend/raw_audio.cc
@@ -32,7 +32,7 @@ void RawAudioCache::Accept(const VectorBase<BaseFloat>& waves) {
         ready_feed_condition_.wait(lock);
     }
     for (size_t idx = 0; idx < waves.Dim(); ++idx) {
-        int32 buffer_idx = (idx + start_) % ring_buffer_.size(); 
+        int32 buffer_idx = (idx + start_) % ring_buffer_.size();
         ring_buffer_[buffer_idx] = waves(idx);
     }
     data_length_ += waves.Dim();
@@ -44,7 +44,8 @@ bool RawAudioCache::Read(Vector<BaseFloat>* waves) {
     std::unique_lock<std::mutex> lock(mutex_);
     while (chunk_size > data_length_) {
         // when audio is empty and no more data feed
-        // ready_read_condition will block in dead lock. so replace with timeout_
+        // ready_read_condition will block in dead lock. so replace with
+        // timeout_
         // ready_read_condition_.wait(lock);
         int32 elapsed = static_cast<int32>(timer.Elapsed() * 1000);
         if (elapsed > timeout_) {
diff --git a/speechx/speechx/frontend/raw_audio.h b/speechx/speechx/frontend/raw_audio.h
index 7726f825b..ce75c137c 100644
--- a/speechx/speechx/frontend/raw_audio.h
+++ b/speechx/speechx/frontend/raw_audio.h
@@ -35,9 +35,9 @@ class RawAudioCache : public FeatureExtractorInterface {
     }
     virtual bool IsFinished() const { return finished_; }
     virtual void Reset() {
-      start_ = 0;
-      data_length_ = 0;
-      finished_ = false;
+        start_ = 0;
+        data_length_ = 0;
+        finished_ = false;
     }
 
   private:
@@ -72,9 +72,7 @@ class RawDataCache : public FeatureExtractorInterface {
     virtual void SetFinished() { finished_ = true; }
     virtual bool IsFinished() const { return finished_; }
     void SetDim(int32 dim) { dim_ = dim; }
-    virtual void Reset() {
-      finished_ = true;
-    }
+    virtual void Reset() { finished_ = true; }
 
   private:
     kaldi::Vector<kaldi::BaseFloat> data_;
diff --git a/speechx/speechx/nnet/decodable-itf.h b/speechx/speechx/nnet/decodable-itf.h
index 37c3007b3..8e9a5a72a 100644
--- a/speechx/speechx/nnet/decodable-itf.h
+++ b/speechx/speechx/nnet/decodable-itf.h
@@ -1,3 +1,17 @@
+// Copyright (c) 2022 PaddlePaddle Authors. All Rights Reserved.
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
 // itf/decodable-itf.h
 
 // Copyright 2009-2011  Microsoft Corporation;  Saarland University;
@@ -42,8 +56,10 @@ namespace kaldi {
 
     For online decoding, where the features are coming in in real time, it is
     important to understand the IsLastFrame() and NumFramesReady() functions.
-    There are two ways these are used: the old online-decoding code, in ../online/,
-    and the new online-decoding code, in ../online2/.  In the old online-decoding
+    There are two ways these are used: the old online-decoding code, in
+   ../online/,
+    and the new online-decoding code, in ../online2/.  In the old
+   online-decoding
     code, the decoder would do:
     \code{.cc}
     for (int frame = 0; !decodable.IsLastFrame(frame); frame++) {
@@ -52,13 +68,16 @@ namespace kaldi {
     \endcode
    and the call to IsLastFrame would block if the features had not arrived yet.
    The decodable object would have to know when to terminate the decoding.  This
-   online-decoding mode is still supported, it is what happens when you call, for
+   online-decoding mode is still supported, it is what happens when you call,
+   for
    example, LatticeFasterDecoder::Decode().
 
    We realized that this "blocking" mode of decoding is not very convenient
    because it forces the program to be multi-threaded and makes it complex to
-   control endpointing.  In the "new" decoding code, you don't call (for example)
-   LatticeFasterDecoder::Decode(), you call LatticeFasterDecoder::InitDecoding(),
+   control endpointing.  In the "new" decoding code, you don't call (for
+   example)
+   LatticeFasterDecoder::Decode(), you call
+   LatticeFasterDecoder::InitDecoding(),
    and then each time you get more features, you provide them to the decodable
    object, and you call LatticeFasterDecoder::AdvanceDecoding(), which does
    something like this:
@@ -68,7 +87,8 @@ namespace kaldi {
    }
    \endcode
    So the decodable object never has IsLastFrame() called.  For decoding where
-   you are starting with a matrix of features, the NumFramesReady() function will
+   you are starting with a matrix of features, the NumFramesReady() function
+   will
    always just return the number of frames in the file, and IsLastFrame() will
    return true for the last frame.
 
@@ -80,45 +100,54 @@ namespace kaldi {
    frame of the file once we've decided to terminate decoding.
 */
 class DecodableInterface {
- public:
-  /// Returns the log likelihood, which will be negated in the decoder.
-  /// The "frame" starts from zero.  You should verify that NumFramesReady() > frame
-  /// before calling this.
-  virtual BaseFloat LogLikelihood(int32 frame, int32 index) = 0;
-
-  /// Returns true if this is the last frame.  Frames are zero-based, so the
-  /// first frame is zero.  IsLastFrame(-1) will return false, unless the file
-  /// is empty (which is a case that I'm not sure all the code will handle, so
-  /// be careful).  Caution: the behavior of this function in an online setting
-  /// is being changed somewhat.  In future it may return false in cases where
-  /// we haven't yet decided to terminate decoding, but later true if we decide
-  /// to terminate decoding.  The plan in future is to rely more on
-  /// NumFramesReady(), and in future, IsLastFrame() would always return false
-  /// in an online-decoding setting, and would only return true in a
-  /// decoding-from-matrix setting where we want to allow the last delta or LDA
-  /// features to be flushed out for compatibility with the baseline setup.
-  virtual bool IsLastFrame(int32 frame) const = 0;
-
-  /// The call NumFramesReady() will return the number of frames currently available
-  /// for this decodable object.  This is for use in setups where you don't want the
-  /// decoder to block while waiting for input.  This is newly added as of Jan 2014,
-  /// and I hope, going forward, to rely on this mechanism more than IsLastFrame to
-  /// know when to stop decoding.
-  virtual int32 NumFramesReady() const {
-    KALDI_ERR << "NumFramesReady() not implemented for this decodable type.";
-    return -1;
-  }
-
-  /// Returns the number of states in the acoustic model
-  /// (they will be indexed one-based, i.e. from 1 to NumIndices();
-  /// this is for compatibility with OpenFst).
-  virtual int32 NumIndices() const = 0;
-
-  virtual bool FrameLogLikelihood(int32 frame, 
-                                  std::vector<kaldi::BaseFloat>* likelihood) = 0;
-
-
-  virtual ~DecodableInterface() {}
+  public:
+    /// Returns the log likelihood, which will be negated in the decoder.
+    /// The "frame" starts from zero.  You should verify that NumFramesReady() >
+    /// frame
+    /// before calling this.
+    virtual BaseFloat LogLikelihood(int32 frame, int32 index) = 0;
+
+    /// Returns true if this is the last frame.  Frames are zero-based, so the
+    /// first frame is zero.  IsLastFrame(-1) will return false, unless the file
+    /// is empty (which is a case that I'm not sure all the code will handle, so
+    /// be careful).  Caution: the behavior of this function in an online
+    /// setting
+    /// is being changed somewhat.  In future it may return false in cases where
+    /// we haven't yet decided to terminate decoding, but later true if we
+    /// decide
+    /// to terminate decoding.  The plan in future is to rely more on
+    /// NumFramesReady(), and in future, IsLastFrame() would always return false
+    /// in an online-decoding setting, and would only return true in a
+    /// decoding-from-matrix setting where we want to allow the last delta or
+    /// LDA
+    /// features to be flushed out for compatibility with the baseline setup.
+    virtual bool IsLastFrame(int32 frame) const = 0;
+
+    /// The call NumFramesReady() will return the number of frames currently
+    /// available
+    /// for this decodable object.  This is for use in setups where you don't
+    /// want the
+    /// decoder to block while waiting for input.  This is newly added as of Jan
+    /// 2014,
+    /// and I hope, going forward, to rely on this mechanism more than
+    /// IsLastFrame to
+    /// know when to stop decoding.
+    virtual int32 NumFramesReady() const {
+        KALDI_ERR
+            << "NumFramesReady() not implemented for this decodable type.";
+        return -1;
+    }
+
+    /// Returns the number of states in the acoustic model
+    /// (they will be indexed one-based, i.e. from 1 to NumIndices();
+    /// this is for compatibility with OpenFst).
+    virtual int32 NumIndices() const = 0;
+
+    virtual bool FrameLogLikelihood(
+        int32 frame, std::vector<kaldi::BaseFloat>* likelihood) = 0;
+
+
+    virtual ~DecodableInterface() {}
 };
 /// @}
 }  // namespace Kaldi
diff --git a/speechx/speechx/nnet/decodable.cc b/speechx/speechx/nnet/decodable.cc
index 3cc07f38b..6c0909ca1 100644
--- a/speechx/speechx/nnet/decodable.cc
+++ b/speechx/speechx/nnet/decodable.cc
@@ -23,10 +23,7 @@ using kaldi::Vector;
 
 Decodable::Decodable(const std::shared_ptr<NnetInterface>& nnet,
                      const std::shared_ptr<FeatureExtractorInterface>& frontend)
-    : frontend_(frontend),
-      nnet_(nnet),
-      frame_offset_(0),
-      frames_ready_(0) {}
+    : frontend_(frontend), nnet_(nnet), frame_offset_(0), frames_ready_(0) {}
 
 void Decodable::Acceptlikelihood(const Matrix<BaseFloat>& likelihood) {
     frames_ready_ += likelihood.NumRows();
@@ -83,7 +80,7 @@ void Decodable::Reset() {
     frontend_->Reset();
     nnet_->Reset();
     frame_offset_ = 0;
-    frames_ready_ = 0; 
+    frames_ready_ = 0;
 }
 
 }  // namespace ppspeech
\ No newline at end of file
diff --git a/utils/generate_infer_yaml.py b/utils/generate_infer_yaml.py
index b8a797ad9..02ff262b0 100644
--- a/utils/generate_infer_yaml.py
+++ b/utils/generate_infer_yaml.py
@@ -148,7 +148,7 @@ def merge_configs(
     for item in remove_train_list:
         try:
             remove_config_part(config, [item])
-        except:
+        except Exception as e:
             print(item + " " + "can not be removed")
 
     # Save the config