format and fix pre-commit (#1120)

dataset/aidatatang_200zh/aidatatang_200zh.py

@@ -25,6 +25,7 @@ import os
 from pathlib import Path
 
 import soundfile
+
 from utils.utility import download
 from utils.utility import unpack
 

dataset/aishell/aishell.py

@@ -25,6 +25,7 @@ import os
 from pathlib import Path
 
 import soundfile
+
 from utils.utility import download
 from utils.utility import unpack
 

dataset/librispeech/librispeech.py

@@ -27,6 +27,7 @@ import os
 from multiprocessing.pool import Pool
 
 import soundfile
+
 from utils.utility import download
 from utils.utility import unpack
 

dataset/mini_librispeech/mini_librispeech.py

@@ -26,6 +26,7 @@ import os
 from multiprocessing.pool import Pool
 
 import soundfile
+
 from utils.utility import download
 from utils.utility import unpack
 

dataset/musan/musan.py

@@ -28,6 +28,7 @@ import json
 import os
 
 import soundfile
+
 from utils.utility import download
 from utils.utility import unpack
 

dataset/rir_noise/rir_noise.py

@@ -28,6 +28,7 @@ import json
 import os
 
 import soundfile
+
 from utils.utility import download
 from utils.utility import unzip
 

dataset/thchs30/thchs30.py

@@ -26,6 +26,7 @@ from multiprocessing.pool import Pool
 from pathlib import Path
 
 import soundfile
+
 from utils.utility import download
 from utils.utility import unpack
 

dataset/timit/timit.py

@@ -27,6 +27,7 @@ import string
 from pathlib import Path
 
 import soundfile
+
 from utils.utility import unzip
 
 URL_ROOT = ""

dataset/voxforge/voxforge.py

@@ -27,6 +27,7 @@ import shutil
 import subprocess
 
 import soundfile
+
 from utils.utility import download_multi
 from utils.utility import getfile_insensitive
 from utils.utility import unpack

docs/source/released_model.md

@@ -75,4 +75,3 @@ PANN | ESC-50 |[pann-esc50]("./examples/esc50/cls0")|[panns_cnn6.tar.gz](https:/
 | [Ds2 Offline Aishell model](https://deepspeech.bj.bcebos.com/mandarin_models/aishell_model_v1.8_to_v2.x.tar.gz) |        Aishell Dataset         | Char-based  | 234 MB | 2 Conv + 3 bidirectional GRU layers                | 0.0804 | -      | 151 h           |
 | [Ds2 Offline Librispeech model](https://deepspeech.bj.bcebos.com/eng_models/librispeech_v1.8_to_v2.x.tar.gz) |      Librispeech Dataset       | Word-based  | 307 MB | 2 Conv + 3 bidirectional sharing weight RNN layers | -      | 0.0685 | 960 h           |
 | [Ds2 Offline Baidu en8k model](https://deepspeech.bj.bcebos.com/eng_models/baidu_en8k_v1.8_to_v2.x.tar.gz) | Baidu Internal English Dataset | Word-based  | 273 MB | 2 Conv + 3 bidirectional GRU layers                | -      | 0.0541 | 8628 h          |
-

paddlespeech/cli/__init__.py

@@ -11,8 +11,8 @@
 # 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.
-
 import _locale
+
 from .asr import ASRExecutor
 from .base_commands import BaseCommand
 from .base_commands import HelpCommand

paddlespeech/cls/exps/panns/deploy/predict.py

@@ -16,10 +16,11 @@ import os
 
 import numpy as np
 from paddle import inference
+from scipy.special import softmax
+
 from paddleaudio.backends import load as load_audio
 from paddleaudio.datasets import ESC50
 from paddleaudio.features import melspectrogram
-from scipy.special import softmax
 
 # yapf: disable
 parser = argparse.ArgumentParser()

paddlespeech/cls/exps/panns/export_model.py

@@ -15,8 +15,8 @@ import argparse
 import os
 
 import paddle
-from paddleaudio.datasets import ESC50
 
+from paddleaudio.datasets import ESC50
 from paddlespeech.cls.models import cnn14
 from paddlespeech.cls.models import SoundClassifier
 

paddlespeech/cls/exps/panns/predict.py

@@ -16,11 +16,11 @@ import argparse
 import numpy as np
 import paddle
 import paddle.nn.functional as F
+
 from paddleaudio.backends import load as load_audio
 from paddleaudio.datasets import ESC50
 from paddleaudio.features import LogMelSpectrogram
 from paddleaudio.features import melspectrogram
-
 from paddlespeech.cls.models import cnn14
 from paddlespeech.cls.models import SoundClassifier
 

paddlespeech/cls/exps/panns/train.py

@@ -15,11 +15,11 @@ import argparse
 import os
 
 import paddle
+
 from paddleaudio.datasets import ESC50
 from paddleaudio.features import LogMelSpectrogram
 from paddleaudio.utils import logger
 from paddleaudio.utils import Timer
-
 from paddlespeech.cls.models import cnn14
 from paddlespeech.cls.models import SoundClassifier
 

paddlespeech/cls/models/panns/panns.py

@@ -15,6 +15,7 @@ import os
 
 import paddle.nn as nn
 import paddle.nn.functional as F
+
 from paddleaudio.utils.download import load_state_dict_from_url
 from paddleaudio.utils.env import MODEL_HOME
 

paddlespeech/s2t/frontend/audio.py

@@ -356,7 +356,7 @@ class AudioSegment():
         # sox, slow
         try:
             import soxbindings as sox
-        except:
+        except ImportError:
             try:
                 from paddlespeech.s2t.utils import dynamic_pip_install
                 package = "sox"
@@ -364,8 +364,9 @@ class AudioSegment():
                 package = "soxbindings"
                 dynamic_pip_install.install(package)
                 import soxbindings as sox
-            except:
-                raise RuntimeError("Can not install soxbindings on your system." )
+            except Exception:
+                raise RuntimeError(
+                    "Can not install soxbindings on your system.")
 
         tfm = sox.Transformer()
         tfm.set_globals(multithread=False)

paddlespeech/s2t/frontend/utility.py

@@ -102,9 +102,11 @@ def read_manifest(
     with jsonlines.open(manifest_path, 'r') as reader:
         for json_data in reader:
             feat_len = json_data["input"][0]["shape"][
-                0] if "input" in json_data and "shape" in json_data["input"][0] else 1.0
+                0] if "input" in json_data and "shape" in json_data["input"][
+                    0] else 1.0
             token_len = json_data["output"][0]["shape"][
-                0] if "output" in json_data and "shape" in json_data["output"][0] else 1.0
+                0] if "output" in json_data and "shape" in json_data["output"][
+                    0] else 1.0
             conditions = [
                 feat_len >= min_input_len,
                 feat_len <= max_input_len,

paddlespeech/s2t/io/sampler.py

@@ -20,13 +20,13 @@ from paddle.io import DistributedBatchSampler
 
 from paddlespeech.s2t.utils.log import Log
 
+logger = Log(__name__).getlog()
+
 __all__ = [
     "SortagradDistributedBatchSampler",
     "SortagradBatchSampler",
 ]
 
-logger = Log(__name__).getlog()
-
 
 def _batch_shuffle(indices, batch_size, epoch, clipped=False):
     """Put similarly-sized instances into minibatches for better efficiency

paddlespeech/s2t/models/ds2/__init__.py

@@ -17,11 +17,11 @@ from paddlespeech.s2t.utils import dynamic_pip_install
 
 try:
     import swig_decoders
-except:
+except ImportError:
     try:
         package_name = 'paddlespeech_ctcdecoders'
         dynamic_pip_install.install(package_name)
-    except:
+    except Exception:
         raise RuntimeError(
             "Can not install package paddlespeech_ctcdecoders on your system. \
                 The DeepSpeech2 model is not supported for your system")

paddlespeech/s2t/models/ds2_online/__init__.py

@@ -17,11 +17,11 @@ from paddlespeech.s2t.utils import dynamic_pip_install
 
 try:
     import swig_decoders
-except:
+except ImportError:
     try:
         package_name = 'paddlespeech_ctcdecoders'
         dynamic_pip_install.install(package_name)
-    except:
+    except Exception:
         raise RuntimeError(
             "Can not install package paddlespeech_ctcdecoders on your system. \
                 The DeepSpeech2 model is not supported for your system")

paddlespeech/s2t/modules/ctc.py

@@ -28,7 +28,7 @@ try:
     from paddlespeech.s2t.decoders.ctcdecoder.swig_wrapper import ctc_beam_search_decoder_batch  # noqa: F401
     from paddlespeech.s2t.decoders.ctcdecoder.swig_wrapper import ctc_greedy_decoder  # noqa: F401
     from paddlespeech.s2t.decoders.ctcdecoder.swig_wrapper import Scorer  # noqa: F401
-except:
+except ImportError:
     try:
         from paddlespeech.s2t.utils import dynamic_pip_install
         package_name = 'paddlespeech_ctcdecoders'

paddlespeech/s2t/training/trainer.py

@@ -221,6 +221,8 @@ class Trainer():
         if hasattr(self.train_loader, "batch_sampler"):
             batch_sampler = self.train_loader.batch_sampler
             if isinstance(batch_sampler, paddle.io.DistributedBatchSampler):
+                logger.debug(
+                    f"train_loader.batch_sample set epoch: {self.epoch}")
                 batch_sampler.set_epoch(self.epoch)
 
     def before_train(self):

paddlespeech/s2t/transform/perturb.py

@@ -147,7 +147,7 @@ class SpeedPerturbationSox():
 
         try:
             import soxbindings as sox
-        except:
+        except ImportError:
             try:
                 from paddlespeech.s2t.utils import dynamic_pip_install
                 package = "sox"
@@ -155,8 +155,10 @@ class SpeedPerturbationSox():
                 package = "soxbindings"
                 dynamic_pip_install.install(package)
                 import soxbindings as sox
-            except:
-                raise RuntimeError("Can not install soxbindings on your system." )
+            except Exception:
+                raise RuntimeError(
+                    "Can not install soxbindings on your system.")
+        self.sox = sox
 
         if utt2ratio is not None:
             self.utt2ratio = {}
@@ -200,7 +202,7 @@ class SpeedPerturbationSox():
         else:
             ratio = self.state.uniform(self.lower, self.upper)
 
-        tfm = sox.Transformer()
+        tfm = self.sox.Transformer()
         tfm.set_globals(multithread=False)
         tfm.speed(ratio)
         y = tfm.build_array(input_array=x, sample_rate_in=self.sr)

utils/manifest_key_value.py

@@ -5,6 +5,7 @@ import functools
 from pathlib import Path
 
 import jsonlines
+
 from utils.utility import add_arguments
 from utils.utility import print_arguments