merge deepspeech, parakeet and text_processing into paddlespeech

.mergify.yml

@@ -41,13 +41,13 @@ pull_request_rules:
  remove: ["conflicts"]
  - name: "auto add label=S2T"
  conditions:
- - files~=^deepspeech/
+ - files~=^paddlespeech/s2t/
  actions:
  label:
  add: ["S2T"]
  - name: "auto add label=T2S"
  conditions:
- - files~=^parakeet/
+ - files~=^paddlespeech/t2s/
  actions:
  label:
  add: ["T2S"]
@@ -59,7 +59,7 @@ pull_request_rules:
  add: ["Audio"]
  - name: "auto add label=TextProcess"
  conditions:
- - files~=^text_processing/
+ - files~=^paddlespeech/text/
  actions:
  label:
  add: ["TextProcess"]

docs/source/asr/models_introduction.md

@@ -61,15 +61,15 @@ For feature extraction, three methods are implemented, which are linear (FFT wit
 Currently, the released deepspeech2 online model use the linear feature extraction method.
 ```
 The code for feature extraction
-vi deepspeech/frontend/featurizer/audio_featurizer.py
+vi paddlespeech/s2t/frontend/featurizer/audio_featurizer.py
 ```
 
 ### Encoder
 The encoder is composed of two 2D convolution subsampling layers and a number of stacked single direction rnn layers. The 2D convolution subsampling layers extract feature representation from the raw audio feature and reduce the length of audio feature at the same time. After passing through the convolution subsampling layers, then the feature representation are input into the stacked rnn layers. For the stacked rnn layers, LSTM cell and GRU cell are provided to use. Adding one fully connected (fc) layer after the stacked rnn layers is optional. If the number of stacked rnn layers is less than 5, adding one fc layer after stacked rnn layers is recommand.
 
 The code of Encoder is in:
 ```
-vi deepspeech/models/ds2_online/deepspeech2.py
+vi paddlespeech/s2t/models/ds2_online/deepspeech2.py
 ```
 
 ### Decoder
@@ -78,9 +78,9 @@ To got the character possibilities of each frame, the feature representation of
 The code of the decoder is in:
 ```
 # The code of constructing the decoder in model
-vi deepspeech/models/ds2_online/deepspeech2.py
+vi paddlespeech/s2t/models/ds2_online/deepspeech2.py
 # The code of CTC Decoder
-vi deepspeech/modules/ctc.py
+vi paddlespeech/s2t/modules/ctc.py
 ```
 
 ### Training Process
@@ -169,7 +169,7 @@ For data preparation and decoder, the deepspeech2 offline model is same with the
 
 The code of encoder and decoder for deepspeech2 offline model is in:
 ```
-vi deepspeech/models/ds2/deepspeech2.py
+vi paddlespeech/s2t/models/ds2/deepspeech2.py
 ```
 
 The training process and testing process of deepspeech2 offline model is very similary to deepspeech2 online model.

docs/source/conf.py

@@ -28,8 +28,8 @@
 # -- Project information -----------------------------------------------------
 
 project = 'paddle speech'
-copyright = '2021, Deepspeech-developers'
-author = 'Deepspeech-developers'
+copyright = '2021, paddlespeech-developers'
+author = 'paddlespeech-developers'
 
 # The full version, including alpha/beta/rc tags
 release = '2.1'

docs/source/index.rst

@@ -1,7 +1,7 @@
-Welcome to paddle Deepspeech documentation !
+Welcome to paddle PaddleSpeech documentation !
 ==============================================
 
-**Deepspeech** is a Speech toolkits implemented by paddlepaddle.
+**PaddleSpeech** is a Speech toolkits implemented by paddlepaddle.
 
 
 Contents

docs/source/tts/advanced_usage.md

@@ -67,7 +67,7 @@ There are two common ways to define a model which consists of several modules.
  ```
  When a model is a complicated and made up of several components, each of which has a separate functionality, and can be replaced by other components with the same functionality, we prefer to define it in this way.
 
-In the directory structure of PaddleSpeech TTS, modules with high reusability are placed in `parakeet.modules`, but models for specific tasks are placed in `parakeet.models`. When developing a new model, developers need to consider the feasibility of splitting the modules, and the degree of generality of the modules, and place them in appropriate directories.
+In the directory structure of PaddleSpeech TTS, modules with high reusability are placed in `paddlespeech.t2s.modules`, but models for specific tasks are placed in `paddlespeech.t2s.models`. When developing a new model, developers need to consider the feasibility of splitting the modules, and the degree of generality of the modules, and place them in appropriate directories.
 
 ## PaddleSpeech TTS's Data Components
 Another critical componnet for a deep learning project is data.
@@ -93,7 +93,7 @@ Then we need to select a format for saving metadata to the hard disk. There are
 
 Meanwhile, `cache` is added here, and a multi-process Manager is used to share memory between multiple processes. When `num_workers` is used, it is guaranteed that each sub process will not cache a copy.
 
-The implementation of `DataTable` can be found in `parakeet/datasets/data_table.py`.
+The implementation of `DataTable` can be found in `paddlespeech/t2s/datasets/data_table.py`.
 ```python
 class DataTable(Dataset):
  """Dataset to load and convert data for general purpose.
@@ -179,9 +179,9 @@ We think this method is a little ugly. We prefer to return the necessary informa
 
 It takes advantage of the globality of Python's module level variables and the effect of context manager.
 
-There is a module level variable in `parakeet/training/reporter.py` `OBSERVATIONS`，which is a `Dict` to store key-value.
+There is a module level variable in `paddlespeech/t2s/training/reporter.py` `OBSERVATIONS`，which is a `Dict` to store key-value.
 ```python
-# parakeet/training/reporter.py
+# paddlespeech/t2s/training/reporter.py
 
 @contextlib.contextmanager
 def scope(observations):

docs/source/tts/quick_start.md

@@ -102,9 +102,9 @@ import numpy as np
 import paddle
 import yaml
 from yacs.config import CfgNode
-from parakeet.models.fastspeech2 import FastSpeech2
-from parakeet.models.fastspeech2 import FastSpeech2Inference
-from parakeet.modules.normalizer import ZScore
+from paddlespeech.t2s.models.fastspeech2 import FastSpeech2
+from paddlespeech.t2s.models.fastspeech2 import FastSpeech2Inference
+from paddlespeech.t2s.modules.normalizer import ZScore
 # examples/fastspeech2/baker/frontend.py
 from frontend import Frontend
 
@@ -161,9 +161,9 @@ import paddle
 import soundfile as sf
 import yaml
 from yacs.config import CfgNode
-from parakeet.models.parallel_wavegan import PWGGenerator
-from parakeet.models.parallel_wavegan import PWGInference
-from parakeet.modules.normalizer import ZScore
+from paddlespeech.t2s.models.parallel_wavegan import PWGGenerator
+from paddlespeech.t2s.models.parallel_wavegan import PWGInference
+from paddlespeech.t2s.modules.normalizer import ZScore
 
 # load the pretrained model
 checkpoint_dir = Path("parallel_wavegan_baker_ckpt_0.4")

examples/aishell/s0/path.sh

@@ -11,4 +11,4 @@ export PYTHONPATH=${MAIN_ROOT}:${PYTHONPATH}
 export LD_LIBRARY_PATH=${LD_LIBRARY_PATH}:/usr/local/lib/
 
 MODEL=deepspeech2
-export BIN_DIR=${MAIN_ROOT}/deepspeech/exps/${MODEL}/bin
+export BIN_DIR=${MAIN_ROOT}/paddlespeech/s2t/exps/${MODEL}/bin

examples/aishell/s1/path.sh

@@ -12,7 +12,7 @@ export LD_LIBRARY_PATH=${LD_LIBRARY_PATH}:/usr/local/lib/
 
 # model exp
 MODEL=u2
-export BIN_DIR=${MAIN_ROOT}/deepspeech/exps/${MODEL}/bin
+export BIN_DIR=${MAIN_ROOT}/paddlespeech/s2t/exps/${MODEL}/bin
 
 
 # srilm

examples/aishell3/tts3/path.sh

@@ -10,4 +10,4 @@ export PYTHONIOENCODING=UTF-8
 export PYTHONPATH=${MAIN_ROOT}:${PYTHONPATH}
 
 MODEL=fastspeech2
-export BIN_DIR=${MAIN_ROOT}/parakeet/exps/${MODEL}
+export BIN_DIR=${MAIN_ROOT}/paddlespeech/t2s/exps/${MODEL}

examples/aishell3/vc0/README.md

@@ -39,7 +39,7 @@ There are silence in the edge of AISHELL-3's wavs, and the audio amplitude is ve
 
 We use Montreal Force Aligner 1.0. The label in aishell3 include pinyin，so the lexicon we provided to MFA is pinyin rather than Chinese characters. And the prosody marks(`$` and `%`) need to be removed. You shoud preprocess the dataset into the format which MFA needs, the texts have the same name with wavs and have the suffix `.lab`.
 
-We use [lexicon.txt](https://github.com/PaddlePaddle/DeepSpeech/blob/develop/parakeet/exps/voice_cloning/tacotron2_ge2e/lexicon.txt) as the lexicon.
+We use [lexicon.txt](https://github.com/PaddlePaddle/DeepSpeech/blob/develop/paddlespeech/t2s/exps/voice_cloning/tacotron2_ge2e/lexicon.txt) as the lexicon.
 
 You can download the alignment results from here [alignment_aishell3.tar.gz](https://paddlespeech.bj.bcebos.com/Parakeet/alignment_aishell3.tar.gz), or train your own MFA model reference to [use_mfa example](https://github.com/PaddlePaddle/DeepSpeech/tree/develop/examples/other/use_mfa) (use MFA1.x now) of our repo.
 

examples/aishell3/vc0/path.sh

@@ -10,4 +10,4 @@ export PYTHONIOENCODING=UTF-8
 export PYTHONPATH=${MAIN_ROOT}:${PYTHONPATH}
 
 MODEL=voice_cloning/tacotron2_ge2e
-export BIN_DIR=${MAIN_ROOT}/parakeet/exps/${MODEL}
+export BIN_DIR=${MAIN_ROOT}/paddlespeech/t2s/exps/${MODEL}

examples/callcenter/s1/path.sh

@@ -12,4 +12,4 @@ export LD_LIBRARY_PATH=${LD_LIBRARY_PATH}:/usr/local/lib/
 
 
 MODEL=u2
-export BIN_DIR=${MAIN_ROOT}/deepspeech/exps/${MODEL}/bin
+export BIN_DIR=${MAIN_ROOT}/paddlespeech/s2t/exps/${MODEL}/bin

examples/csmsc/tts2/path.sh

@@ -10,4 +10,4 @@ export PYTHONIOENCODING=UTF-8
 export PYTHONPATH=${MAIN_ROOT}:${PYTHONPATH}
 
 MODEL=speedyspeech
-export BIN_DIR=${MAIN_ROOT}/parakeet/exps/${MODEL}
+export BIN_DIR=${MAIN_ROOT}/paddlespeech/t2s/exps/${MODEL}

examples/csmsc/tts3/path.sh

@@ -10,4 +10,4 @@ export PYTHONIOENCODING=UTF-8
 export PYTHONPATH=${MAIN_ROOT}:${PYTHONPATH}
 
 MODEL=fastspeech2
-export BIN_DIR=${MAIN_ROOT}/parakeet/exps/${MODEL}
+export BIN_DIR=${MAIN_ROOT}/paddlespeech/t2s/exps/${MODEL}

examples/csmsc/voc1/path.sh

@@ -10,4 +10,4 @@ export PYTHONIOENCODING=UTF-8
 export PYTHONPATH=${MAIN_ROOT}:${PYTHONPATH}
 
 MODEL=parallelwave_gan
-export BIN_DIR=${MAIN_ROOT}/parakeet/exps/gan_vocoder/${MODEL}
+export BIN_DIR=${MAIN_ROOT}/paddlespeech/t2s/exps/gan_vocoder/${MODEL}

examples/csmsc/voc3/path.sh

@@ -10,4 +10,4 @@ export PYTHONIOENCODING=UTF-8
 export PYTHONPATH=${MAIN_ROOT}:${PYTHONPATH}
 
 MODEL=multi_band_melgan
-export BIN_DIR=${MAIN_ROOT}/parakeet/exps/gan_vocoder/${MODEL}
+export BIN_DIR=${MAIN_ROOT}/paddlespeech/t2s/exps/gan_vocoder/${MODEL}

examples/librispeech/s0/path.sh

@@ -12,4 +12,4 @@ export LD_LIBRARY_PATH=${LD_LIBRARY_PATH}:/usr/local/lib/
 
 
 MODEL=deepspeech2
-export BIN_DIR=${MAIN_ROOT}/deepspeech/exps/${MODEL}/bin
+export BIN_DIR=${MAIN_ROOT}/paddlespeech/s2t/exps/${MODEL}/bin

examples/librispeech/s1/path.sh

@@ -12,4 +12,4 @@ export LD_LIBRARY_PATH=${LD_LIBRARY_PATH}:/usr/local/lib/
 
 
 MODEL=u2
-export BIN_DIR=${MAIN_ROOT}/deepspeech/exps/${MODEL}/bin
+export BIN_DIR=${MAIN_ROOT}/paddlespeech/s2t/exps/${MODEL}/bin

examples/librispeech/s2/path.sh

@@ -12,7 +12,7 @@ export LD_LIBRARY_PATH=${LD_LIBRARY_PATH}:/usr/local/lib/
 
 
 MODEL=u2_kaldi
-export BIN_DIR=${MAIN_ROOT}/deepspeech/exps/${MODEL}/bin
+export BIN_DIR=${MAIN_ROOT}/paddlespeech/s2t/exps/${MODEL}/bin
 
 # srilm
 export LIBLBFGS=${MAIN_ROOT}/tools/liblbfgs-1.10

examples/ljspeech/tts0/path.sh

@@ -10,4 +10,4 @@ export PYTHONIOENCODING=UTF-8
 export PYTHONPATH=${MAIN_ROOT}:${PYTHONPATH}
 
 MODEL=tacotron2
-export BIN_DIR=${MAIN_ROOT}/parakeet/exps/${MODEL}
+export BIN_DIR=${MAIN_ROOT}/paddlespeech/t2s/exps/${MODEL}

examples/ljspeech/tts1/path.sh

@@ -10,4 +10,4 @@ export PYTHONIOENCODING=UTF-8
 export PYTHONPATH=${MAIN_ROOT}:${PYTHONPATH}
 
 MODEL=transformer_tts
-export BIN_DIR=${MAIN_ROOT}/parakeet/exps/${MODEL}
+export BIN_DIR=${MAIN_ROOT}/paddlespeech/t2s/exps/${MODEL}

examples/ljspeech/tts3/path.sh

@@ -10,4 +10,4 @@ export PYTHONIOENCODING=UTF-8
 export PYTHONPATH=${MAIN_ROOT}:${PYTHONPATH}
 
 MODEL=fastspeech2
-export BIN_DIR=${MAIN_ROOT}/parakeet/exps/${MODEL}
+export BIN_DIR=${MAIN_ROOT}/paddlespeech/t2s/exps/${MODEL}

examples/ljspeech/voc0/path.sh

@@ -10,4 +10,4 @@ export PYTHONIOENCODING=UTF-8
 export PYTHONPATH=${MAIN_ROOT}:${PYTHONPATH}
 
 MODEL=waveflow
-export BIN_DIR=${MAIN_ROOT}/parakeet/exps/${MODEL}
+export BIN_DIR=${MAIN_ROOT}/paddlespeech/t2s/exps/${MODEL}

examples/ljspeech/voc1/path.sh

@@ -10,4 +10,4 @@ export PYTHONIOENCODING=UTF-8
 export PYTHONPATH=${MAIN_ROOT}:${PYTHONPATH}
 
 MODEL=parallelwave_gan
-export BIN_DIR=${MAIN_ROOT}/parakeet/exps/gan_vocoder/${MODEL}
+export BIN_DIR=${MAIN_ROOT}/paddlespeech/t2s/exps/gan_vocoder/${MODEL}

examples/other/1xt2x/src_deepspeech2x/__init__.py

@@ -21,7 +21,7 @@
 from paddle.fluid import core
 from paddle.nn import functional as F
 
-from deepspeech.utils.log import Log
+from paddlespeech.s2t.utils.log import Log
 
 #TODO(Hui Zhang): remove fluid import
 logger = Log(__name__).getlog()

examples/other/1xt2x/src_deepspeech2x/bin/test.py

@@ -14,9 +14,9 @@
 """Evaluation for DeepSpeech2 model."""
 from src_deepspeech2x.test_model import DeepSpeech2Tester as Tester
 
-from deepspeech.exps.deepspeech2.config import get_cfg_defaults
-from deepspeech.training.cli import default_argument_parser
-from deepspeech.utils.utility import print_arguments
+from paddlespeech.s2t.exps.deepspeech2.config import get_cfg_defaults
+from paddlespeech.s2t.training.cli import default_argument_parser
+from paddlespeech.s2t.utils.utility import print_arguments
 
 
 def main_sp(config, args):

examples/other/1xt2x/src_deepspeech2x/models/ds2/deepspeech2.py

@@ -19,11 +19,11 @@
 from src_deepspeech2x.models.ds2.rnn import RNNStack
 from yacs.config import CfgNode
 
-from deepspeech.models.ds2.conv import ConvStack
-from deepspeech.modules.ctc import CTCDecoder
-from deepspeech.utils import layer_tools
-from deepspeech.utils.checkpoint import Checkpoint
-from deepspeech.utils.log import Log
+from paddlespeech.s2t.models.ds2.conv import ConvStack
+from paddlespeech.s2t.modules.ctc import CTCDecoder
+from paddlespeech.s2t.utils import layer_tools
+from paddlespeech.s2t.utils.checkpoint import Checkpoint
+from paddlespeech.s2t.utils.log import Log
 logger = Log(__name__).getlog()
 
 __all__ = ['DeepSpeech2Model', 'DeepSpeech2InferModel']

examples/other/1xt2x/src_deepspeech2x/models/ds2/rnn.py

@@ -18,9 +18,9 @@
 from paddle.nn import functional as F
 from paddle.nn import initializer as I
 
-from deepspeech.modules.activation import brelu
-from deepspeech.modules.mask import make_non_pad_mask
-from deepspeech.utils.log import Log
+from paddlespeech.s2t.modules.activation import brelu
+from paddlespeech.s2t.modules.mask import make_non_pad_mask
+from paddlespeech.s2t.utils.log import Log
 logger = Log(__name__).getlog()
 
 __all__ = ['RNNStack']

examples/other/1xt2x/src_deepspeech2x/test_model.py

@@ -26,19 +26,19 @@
 from src_deepspeech2x.models.ds2 import DeepSpeech2Model
 from yacs.config import CfgNode
 
-from deepspeech.frontend.featurizer.text_featurizer import TextFeaturizer
-from deepspeech.io.collator import SpeechCollator
-from deepspeech.io.dataset import ManifestDataset
-from deepspeech.io.sampler import SortagradBatchSampler
-from deepspeech.io.sampler import SortagradDistributedBatchSampler
-from deepspeech.models.ds2_online import DeepSpeech2InferModelOnline
-from deepspeech.models.ds2_online import DeepSpeech2ModelOnline
-from deepspeech.training.gradclip import ClipGradByGlobalNormWithLog
-from deepspeech.training.trainer import Trainer
-from deepspeech.utils import error_rate
-from deepspeech.utils import layer_tools
-from deepspeech.utils import mp_tools
-from deepspeech.utils.log import Log
+from paddlespeech.s2t.frontend.featurizer.text_featurizer import TextFeaturizer
+from paddlespeech.s2t.io.collator import SpeechCollator
+from paddlespeech.s2t.io.dataset import ManifestDataset
+from paddlespeech.s2t.io.sampler import SortagradBatchSampler
+from paddlespeech.s2t.io.sampler import SortagradDistributedBatchSampler
+from paddlespeech.s2t.models.ds2_online import DeepSpeech2InferModelOnline
+from paddlespeech.s2t.models.ds2_online import DeepSpeech2ModelOnline
+from paddlespeech.s2t.training.gradclip import ClipGradByGlobalNormWithLog
+from paddlespeech.s2t.training.trainer import Trainer
+from paddlespeech.s2t.utils import error_rate
+from paddlespeech.s2t.utils import layer_tools
+from paddlespeech.s2t.utils import mp_tools
+from paddlespeech.s2t.utils.log import Log
 
 logger = Log(__name__).getlog()
 

examples/other/ge2e/path.sh

@@ -10,4 +10,4 @@ export PYTHONIOENCODING=UTF-8
 export PYTHONPATH=${MAIN_ROOT}:${PYTHONPATH}
 
 MODEL=ge2e
-export BIN_DIR=${MAIN_ROOT}/parakeet/exps/${MODEL}
+export BIN_DIR=${MAIN_ROOT}/paddlespeech/t2s/exps/${MODEL}

examples/other/text_frontend/test_g2p.py

@@ -15,8 +15,8 @@
 import re
 from pathlib import Path
 
-from parakeet.frontend.zh_frontend import Frontend as zhFrontend
-from parakeet.utils.error_rate import word_errors
+from paddlespeech.t2s.frontend.zh_frontend import Frontend as zhFrontend
+from paddlespeech.t2s.utils.error_rate import word_errors
 
 SILENCE_TOKENS = {"sp", "sil", "sp1", "spl"}
 

examples/other/text_frontend/test_textnorm.py

@@ -15,8 +15,8 @@
 import re
 from pathlib import Path
 
-from parakeet.frontend.zh_normalization.text_normlization import TextNormalizer
-from parakeet.utils.error_rate import char_errors
+from paddlespeech.t2s.frontend.zh_normalization.text_normlization import TextNormalizer
+from paddlespeech.t2s.utils.error_rate import char_errors
 
 
 # delete english characters

examples/ted_en_zh/t0/path.sh

@@ -12,4 +12,4 @@ export LD_LIBRARY_PATH=${LD_LIBRARY_PATH}:/usr/local/lib/
 
 
 MODEL=u2_st
-export BIN_DIR=${MAIN_ROOT}/deepspeech/exps/${MODEL}/bin
+export BIN_DIR=${MAIN_ROOT}/paddlespeech/s2t/exps/${MODEL}/bin

examples/timit/s1/path.sh

@@ -12,4 +12,4 @@ export LD_LIBRARY_PATH=${LD_LIBRARY_PATH}:/usr/local/lib/
 
 
 MODEL=u2
-export BIN_DIR=${MAIN_ROOT}/deepspeech/exps/${MODEL}/bin
+export BIN_DIR=${MAIN_ROOT}/paddlespeech/s2t/exps/${MODEL}/bin

examples/tiny/s0/path.sh

@@ -11,4 +11,4 @@ export LD_LIBRARY_PATH=${LD_LIBRARY_PATH}:/usr/local/lib/
 
 
 MODEL=deepspeech2
-export BIN_DIR=${MAIN_ROOT}/deepspeech/exps/${MODEL}/bin
+export BIN_DIR=${MAIN_ROOT}/paddlespeech/s2t/exps/${MODEL}/bin