Merge pull request #741 from PaddlePaddle/utils

duration utils

deepspeech/decoders/swig/setup.py

@@ -84,9 +84,8 @@ FILES = glob.glob('kenlm/util/*.cc') \
 FILES += glob.glob('openfst-1.6.3/src/lib/*.cc')
 
 FILES = [
-    fn for fn in FILES
-    if not (fn.endswith('main.cc') or fn.endswith('test.cc') or fn.endswith(
-        'unittest.cc'))
+    fn for fn in FILES if not (fn.endswith('main.cc') or fn.endswith('test.cc')
+                               or fn.endswith('unittest.cc'))
 ]
 
 LIBS = ['stdc++']

deepspeech/models/ds2/__init__.py

@@ -1,7 +1,17 @@
-from .deepspeech2 import DeepSpeech2Model
+# Copyright (c) 2021 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.
 from .deepspeech2 import DeepSpeech2InferModel
+from .deepspeech2 import DeepSpeech2Model
 
 __all__ = ['DeepSpeech2Model', 'DeepSpeech2InferModel']
-
-
-

deepspeech/models/ds2/deepspeech2.py

@@ -19,15 +19,15 @@ from paddle import nn
 from yacs.config import CfgNode
 
 from deepspeech.models.ds2.conv import ConvStack
-from deepspeech.modules.ctc import CTCDecoder
 from deepspeech.models.ds2.rnn import RNNStack
+from deepspeech.modules.ctc import CTCDecoder
 from deepspeech.utils import layer_tools
 from deepspeech.utils.checkpoint import Checkpoint
 from deepspeech.utils.log import Log
 
 logger = Log(__name__).getlog()
 
-__all__ = ['DeepSpeech2Model', 'DeepSpeech2InferMode']
+__all__ = ['DeepSpeech2Model', 'DeepSpeech2InferModel']
 
 
 class CRNNEncoder(nn.Layer):
@@ -117,7 +117,7 @@ class DeepSpeech2Model(nn.Layer):
     :type share_weights: bool
     :return: A tuple of an output unnormalized log probability layer (
              before softmax) and a ctc cost layer.
-    :rtype: tuple of LayerOutput    
+    :rtype: tuple of LayerOutput
     """
 
     @classmethod
@@ -206,10 +206,10 @@ class DeepSpeech2Model(nn.Layer):
 
         config: yacs.config.CfgNode
             model configs
-        
+
         checkpoint_path: Path or str
             the path of pretrained model checkpoint, without extension name
-        
+
         Returns
         -------
         DeepSpeech2Model

examples/callcenter/s1/README.md

@@ -17,4 +17,4 @@
 | conformer | 45.73 M | conf/chunk_conformer.yaml | spec_aug + shift | test | attention | 16, -1 | 2.23287845  | 0.087982 |  
 | conformer | 45.73 M | conf/chunk_conformer.yaml | spec_aug + shift | test | ctc_greedy_search | 16, -1 | 2.23287845  | 0.086962 |  
 | conformer | 45.73 M | conf/chunk_conformer.yaml | spec_aug + shift | test | ctc_prefix_beam_search | 16, -1 | 2.23287845 | 0.086741 |  
-| conformer | 45.73 M | conf/chunk_conformer.yaml | spec_aug + shift | test | attention_rescoring | 16, -1 | 2.23287845 | 0.083495 | 
+| conformer | 45.73 M | conf/chunk_conformer.yaml | spec_aug + shift | test | attention_rescoring | 16, -1 | 2.23287845 | 0.083495 |

examples/librispeech/s1/README.md

@@ -1,5 +1,12 @@
 # LibriSpeech
 
+## Data
+
+| Data Subset | Duration in Seconds |
+| data/manifest.train |  0.83s ~ 29.735s |
+| data/manifest.dev | 1.065 ~ 35.155s |  
+| data/manifest.test-clean | 1.285s ~ 34.955s |
+
 ## Conformer
 
 | Model | Params | Config | Augmentation| Test set | Decode method | Loss | WER |  

examples/librispeech/s1/path.sh

@@ -1,10 +1,10 @@
 export MAIN_ROOT=${PWD}/../../../
 
-export PATH=${MAIN_ROOT}:${MAIN_ROOT}/utils:${PATH}
+export PATH=${MAIN_ROOT}:${PWD}/utils:${PATH}
 export LC_ALL=C
 
 # Use UTF-8 in Python to avoid UnicodeDecodeError when LC_ALL=C
-export PYTHONIOENCODING=UTF-8 
+export PYTHONIOENCODING=UTF-8
 export PYTHONPATH=${MAIN_ROOT}:${PYTHONPATH}
 
 export LD_LIBRARY_PATH=${LD_LIBRARY_PATH}:/usr/local/lib/

examples/librispeech/s1/utils

@@ -0,0 +1 @@
+../../../utils

setup.sh

@@ -10,7 +10,7 @@ fi
 
 if [ -e /etc/lsb-release ];then
     #${SUDO} apt-get update
-    ${SUDO} apt-get install -y vim tig tree sox pkg-config libflac-dev libogg-dev libvorbis-dev libboost-dev swig python3-dev
+    ${SUDO} apt-get install -y jq vim tig tree sox pkg-config libflac-dev libogg-dev libvorbis-dev libboost-dev swig python3-dev
     if [ $? != 0 ]; then
         error_msg "Please using Ubuntu or install pkg-config libflac-dev libogg-dev libvorbis-dev libboost-dev swig python3-dev by user."
         exit -1
@@ -54,10 +54,10 @@ if [ $? != 0 ]; then
     fi
     cd AutoLog
     pip install -r requirements.txt
-    python setup.py install 
+    python setup.py install
     cd ..
     rm -rf AutoLog
-fi 
+fi
 
 # install decoders
 python3 -c "import pkg_resources; pkg_resources.require(\"swig_decoders==1.1\")"

utils/duration_from_maniefst.sh

@@ -0,0 +1,10 @@
+#!/bin/bash
+
+if [ $# == 1 ];then
+    echo "usage: ${0} manifest_file"
+    exit -1
+fi
+
+manifest=$1
+
+jq -S '.feat_shape[0]' ${manifest} | sort -nu