[speechx] refactor speech egs

.pre-commit-config.yaml

@@ -50,13 +50,13 @@ repos:
         entry: bash .pre-commit-hooks/clang-format.hook -i
         language: system
         files: \.(c|cc|cxx|cpp|cu|h|hpp|hxx|cuh|proto)$
-        exclude: (?=speechx/speechx/kaldi|speechx/patch).*(\.cpp|\.cc|\.h|\.py)$
+        exclude: (?=speechx/speechx/kaldi|speechx/patch|speechx/tools/fstbin|speechx/tools/lmbin).*(\.cpp|\.cc|\.h|\.py)$
     -   id: copyright_checker
         name: copyright_checker
         entry: python .pre-commit-hooks/copyright-check.hook
         language: system
         files: \.(c|cc|cxx|cpp|cu|h|hpp|hxx|proto|py)$
-        exclude: (?=third_party|pypinyin|speechx/speechx/kaldi|speechx/patch).*(\.cpp|\.cc|\.h|\.py)$
+        exclude: (?=third_party|pypinyin|speechx/speechx/kaldi|speechx/patch|speechx/tools/fstbin|speechx/tools/lmbin).*(\.cpp|\.cc|\.h|\.py)$
 -   repo: https://github.com/asottile/reorder_python_imports
     rev: v2.4.0
     hooks:

demos/audio_searching/src/audio_search.py

@@ -20,6 +20,7 @@
 from fastapi import FastAPI
 from fastapi import File
 from fastapi import UploadFile
+from logs import LOGGER
 from milvus_helpers import MilvusHelper
 from mysql_helpers import MySQLHelper
 from operations.count import do_count
@@ -31,8 +32,6 @@
 from starlette.requests import Request
 from starlette.responses import FileResponse
 
-from logs import LOGGER
-
 app = FastAPI()
 app.add_middleware(
     CORSMiddleware,

demos/audio_searching/src/encode.py

@@ -12,8 +12,8 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 import numpy as np
-
 from logs import LOGGER
+
 from paddlespeech.cli import VectorExecutor
 
 vector_executor = VectorExecutor()

demos/audio_searching/src/mysql_helpers.py

@@ -20,7 +20,6 @@
 from config import MYSQL_PORT
 from config import MYSQL_PWD
 from config import MYSQL_USER
-
 from logs import LOGGER
 
 

demos/audio_searching/src/operations/count.py

@@ -14,7 +14,6 @@
 import sys
 
 from config import DEFAULT_TABLE
-
 from logs import LOGGER
 
 

demos/audio_searching/src/operations/drop.py

@@ -14,7 +14,6 @@
 import sys
 
 from config import DEFAULT_TABLE
-
 from logs import LOGGER
 
 

demos/audio_searching/src/operations/load.py

@@ -17,7 +17,6 @@
 from config import DEFAULT_TABLE
 from diskcache import Cache
 from encode import get_audio_embedding
-
 from logs import LOGGER
 
 
@@ -27,9 +26,8 @@ def get_audios(path):
     """
     supported_formats = [".wav", ".mp3", ".ogg", ".flac", ".m4a"]
     return [
-        item
-        for sublist in [[os.path.join(dir, file) for file in files]
-                        for dir, _, files in list(os.walk(path))]
+        item for sublist in [[os.path.join(dir, file) for file in files]
+                             for dir, _, files in list(os.walk(path))]
         for item in sublist if os.path.splitext(item)[1] in supported_formats
     ]
 

demos/audio_searching/src/operations/search.py

@@ -17,7 +17,6 @@
 from config import DEFAULT_TABLE
 from config import TOP_K
 from encode import get_audio_embedding
-
 from logs import LOGGER
 
 

demos/audio_searching/src/vpr_search.py

@@ -18,6 +18,7 @@
 from fastapi import FastAPI
 from fastapi import File
 from fastapi import UploadFile
+from logs import LOGGER
 from mysql_helpers import MySQLHelper
 from operations.count import do_count_vpr
 from operations.count import do_get
@@ -30,8 +31,6 @@
 from starlette.requests import Request
 from starlette.responses import FileResponse
 
-from logs import LOGGER
-
 app = FastAPI()
 app.add_middleware(
     CORSMiddleware,

paddleaudio/setup.py

@@ -84,7 +84,7 @@ def remove_version_py(filename='paddleaudio/__init__.py'):
     install_requires=[
         'numpy >= 1.15.0', 'scipy >= 1.0.0', 'resampy >= 0.2.2',
         'soundfile >= 0.9.0', 'colorlog', 'dtaidistance == 2.3.1', 'pathos'
-        ],
+    ],
     extras_require={
         'test': [
             'nose', 'librosa==0.8.1', 'soundfile==0.10.3.post1',

paddlespeech/server/bin/paddlespeech_client.py

@@ -12,15 +12,15 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 import argparse
+import asyncio
 import base64
 import io
 import json
+import logging
 import os
 import random
 import time
 from typing import List
-import logging
-import asyncio
 
 import numpy as np
 import requests
@@ -30,9 +30,9 @@
 from ..util import cli_client_register
 from ..util import stats_wrapper
 from paddlespeech.cli.log import logger
+from paddlespeech.server.tests.asr.online.websocket_client import ASRAudioHandler
 from paddlespeech.server.utils.audio_process import wav2pcm
 from paddlespeech.server.utils.util import wav2base64
-from paddlespeech.server.tests.asr.online.websocket_client import ASRAudioHandler
 
 __all__ = ['TTSClientExecutor', 'ASRClientExecutor', 'CLSClientExecutor']
 
@@ -234,7 +234,8 @@ def __call__(self,
 
 
 @cli_client_register(
-    name='paddlespeech_client.asr_online', description='visit asr online service')
+    name='paddlespeech_client.asr_online',
+    description='visit asr online service')
 class ASRClientExecutor(BaseExecutor):
     def __init__(self):
         super(ASRClientExecutor, self).__init__()

paddlespeech/server/tests/asr/online/web/app.py

@@ -1,22 +1,23 @@
 #!/usr/bin/env python3
 # -*- coding: utf-8 -*-
-
 # Copyright 2021 Mobvoi Inc. All Rights Reserved.
 # Author: [email protected] (Zhendong Peng)
-
 import argparse
 
-from flask import Flask, render_template
+from flask import Flask
+from flask import render_template
 
 parser = argparse.ArgumentParser(description='training your network')
 parser.add_argument('--port', default=19999, type=int, help='port id')
 args = parser.parse_args()
 
 app = Flask(__name__)
 
+
 @app.route('/')
 def index():
     return render_template('index.html')
 
+
 if __name__ == '__main__':
     app.run(host='0.0.0.0', port=args.port, debug=True)

paddlespeech/server/tests/asr/online/web/readme.md

@@ -15,4 +15,4 @@
 在浏览器中输入127.0.0.1:19999 即可看到相关网页Demo。
 
 ![图片](./paddle_web_demo.png)
-
+

paddlespeech/vector/io/dataset.py

@@ -13,6 +13,7 @@
 # limitations under the License.
 from dataclasses import dataclass
 from dataclasses import fields
+
 from paddle.io import Dataset
 
 from paddleaudio import load as load_audio

paddlespeech/vector/io/dataset_from_json.py

@@ -12,9 +12,9 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 import json
-
 from dataclasses import dataclass
 from dataclasses import fields
+
 from paddle.io import Dataset
 
 from paddleaudio import load as load_audio

speechx/examples/CMakeLists.txt

@@ -1,7 +1,4 @@
 cmake_minimum_required(VERSION 3.14 FATAL_ERROR)
 
-add_subdirectory(feat)
-add_subdirectory(nnet)
-add_subdirectory(decoder)
-
-add_subdirectory(glog)
+add_subdirectory(ds2_ol)
+add_subdirectory(dev)

speechx/examples/README.md

@@ -1,17 +1,25 @@
-# Examples
+# Examples for SpeechX
+
+* dev - for speechx developer, using for test.
+* ngram - using to build NGram ARPA lm.
+* ds2_ol - ds2 streaming test under `aishell-1` test dataset.
+ The entrypoint is `ds2_ol/aishell/run.sh`
 
-* glog - glog usage
-* feat - mfcc, linear 
-* nnet - ds2 nn
-* decoder - online decoder to work as offline
 
 ## How to run
 
 `run.sh` is the entry point.
 
-Example to play `decoder`:
+Example to play `ds2_ol`:
 
 ```
-pushd decoder
+pushd ds2_ol/aishell
 bash run.sh
 ```
+
+## Display Model with [Netron](https://github.com/lutzroeder/netron)
+
+```
+pip install netron
+netron exp/deepspeech2_online/checkpoints/avg_1.jit.pdmodel  --port 8022 --host 10.21.55.20
+```