feat: 添加基于http协议的新调用接口

asrt_sdk/Recorder.py

@@ -1,9 +1,33 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+#
+# Copyright 2016-2099 Ailemon.net
+#
+# This file is part of ASRT Speech Recognition Tool Python SDK.
+#
+# ASRT is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+# ASRT is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with ASRT.  If not, see <https://www.gnu.org/licenses/>.
+# ============================================================================
+
+"""
+@author: nl8590687
+ASRT语音识别Python SDK 录音功能库
+"""
 
 import wave
 from pyaudio import PyAudio, paInt16
 import struct
 import threading
-
+import numpy as np
 
 
 class AudioRecorder():
@@ -70,11 +94,19 @@ def SaveAudioToFile(self, filename):
         pass
 
     def GetAudioStream(self):
-        return b"".join(self.__audio_buffers__)
+        bytesStream = b"".join(self.__audio_buffers__)
+        #print(bytesStream[-1000:])
+        #f=open('test0.bin','wb')
+        #f.write(bytesStream)
+        #f.close()
+        return bytesStream
 
     def GetAudioSamples(self):
         audio_bin_serials = self.GetAudioStream()
-        return self.__audio_stream_to_short__(audio_bin_serials)
+        wave_data = np.fromstring(audio_bin_serials, dtype = np.short) # 将声音文件数据转换为数组矩阵形式
+        #print(wave_data[-1000:])
+        return wave_data
+        #return self.__audio_stream_to_short__(audio_bin_serials)
         pass
 
     def __audio_stream_to_short__(self, audio_stream):

asrt_sdk/SpeechRecognizer.py

@@ -1,11 +1,122 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+#
+# Copyright 2016-2099 Ailemon.net
+#
+# This file is part of ASRT Speech Recognition Tool Python SDK.
+#
+# ASRT is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+# ASRT is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with ASRT.  If not, see <https://www.gnu.org/licenses/>.
+# ============================================================================
+
+"""
+@author: nl8590687
+ASRT语音识别Python SDK 语音识别接口调用类库
+"""
 
-from .Recorder import AudioRecorder
 import threading
 import time
 import wave
 import requests
 import numpy as np
 
+from .Recorder import AudioRecorder
+from .utils import *
+
+def get_speech_recognizer(host:str, port:str, protocol:str):
+    '''
+    获取一个ASRT语音识别SDK接口调用实例对象 \\
+    参数：\\
+        host: 主机域名或IP.
+        port: 主机端口号.
+        protocol: 网络协议类型.
+    '''
+    if protocol.lower() == 'http' or protocol.lower() == 'https':
+        return HttpSpeechRecognizer(host, port, protocol)
+    return None
+
+
+class BaseSpeechRecognizer():
+    '''
+    ASRT语音识别SDK接口调用类基类
+    '''
+    def __init__(self, host:str, port:str, protocol:str):
+        self.host = host
+        self.port = port
+        self.protocol = protocol
+
+    def recognite(self, wav_data, frame_rate, channels, byte_width):
+        raise Exception("Method Unimpletment")
+
+    def recognite_speech(self, wav_data, frame_rate, channels, byte_width):
+        raise Exception("Method Unimpletment")
+
+    def recognite_language(self, sequence_pinyin):
+        raise Exception("Method Unimpletment")
+
+    def recognite_file(self, filename):
+        wave_data = read_wav_datas(filename)
+        str_data = wave_data.str_data
+        frame_rate = wave_data.sample_rate
+        channels = wave_data.channels
+        byte_width = wave_data.byte_width
+        return self.recognite(wav_data=str_data,
+                            frame_rate=frame_rate,
+                            channels=channels,
+                            byte_width=byte_width
+                            )
+
+class HttpSpeechRecognizer(BaseSpeechRecognizer):
+    '''
+    ASRT语音识别SDK基于HTTP协议接口调用类 \\
+    参数: \\
+        host: 主机域名或IP.
+        port: 主机端口号.
+        protocol: 网络协议类型.
+        sub_path: 接口所在URL的子路径, 默认为""
+    '''
+    def __init__(self, host:str, port:str, protocol:str, sub_path:str=''):
+        super().__init__(host, port, protocol)
+        if protocol != 'http' and protocol != 'https':
+            raise Exception('Unsupport netword protocol `' + protocol +'`')
+        self._url_ = protocol + '://' + host + ':' + port
+        self.sub_path = sub_path
+
+    def recognite(self, wav_data, frame_rate:int, channels:int, byte_width:int) -> AsrtApiResponse:
+        request_body = AsrtApiSpeechRequest(wav_data, frame_rate, channels, byte_width)
+        headers = {'Content-Type': 'application/json'}
+        response_object = requests.post(self._url_ + self.sub_path + '/all', headers=headers, data=request_body.to_json())
+        response_body_dict = json.loads(response_object.text)
+        response_body = AsrtApiResponse()
+        response_body.from_json(**response_body_dict)
+        return response_body
+
+    def recognite_speech(self, wav_data, frame_rate, channels, byte_width):
+        request_body = AsrtApiSpeechRequest(wav_data, frame_rate, channels, byte_width)
+        headers = {'Content-Type': 'application/json'}
+        response_object = requests.post(self._url_ + self.sub_path + '/speech', headers=headers, data=request_body.to_json())
+        response_body_dict = json.loads(response_object.text)
+        response_body = AsrtApiResponse()
+        response_body.from_json(**response_body_dict)
+        return response_body
+
+    def recognite_language(self, sequence_pinyin):
+        request_body = AsrtApiLanguageRequest(sequence_pinyin)
+        headers = {'Content-Type': 'application/json'}
+        response_object = requests.post(self._url_ + self.sub_path + '/language', headers=headers, data=request_body.to_json())
+        response_body_dict = json.loads(response_object.text)
+        response_body = AsrtApiResponse()
+        response_body.from_json(**response_body_dict)
+        return response_body
 
 class SpeechRecognizer():
     def __init__(self, url_server = 'http://127.0.0.1:20000/', token_client = 'qwertasd'):

asrt_sdk/__init__.py

@@ -19,5 +19,7 @@
 from . import Recorder, SpeechRecognizer
 from .Recorder import AudioRecorder
 from .SpeechRecognizer import SpeechRecognizer
+from .SpeechRecognizer import get_speech_recognizer, HttpSpeechRecognizer
+from .utils import read_wav_datas
 
 __version__ = '1.0.0'

asrt_sdk/utils.py

@@ -0,0 +1,160 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+#
+# Copyright 2016-2099 Ailemon.net
+#
+# This file is part of ASRT Speech Recognition Tool Python SDK.
+#
+# ASRT is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+# ASRT is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with ASRT.  If not, see <https://www.gnu.org/licenses/>.
+# ============================================================================
+
+"""
+@author: nl8590687
+ASRT语音识别Python SDK 基础库模块
+"""
+
+import base64
+import json
+import wave
+import numpy as np
+
+API_STATUS_CODE_OK = 200000 # OK
+API_STATUS_CODE_CLIENT_ERROR = 400000
+API_STATUS_CODE_CLIENT_ERROR_FORMAT = 400001 # 请求数据格式错误
+API_STATUS_CODE_CLIENT_ERROR_FORMAT = 400002 # 请求数据配置不支持
+API_STATUS_CODE_SERVER_ERROR = 500000
+API_STATUS_CODE_SERVER_ERROR_RUNNING = 500001 # 服务器运行中出错
+
+class AsrtApiSpeechRequest:
+    '''
+    ASRT语音识别基于HTTP协议的API接口请求类(声学模型)
+    '''
+    def __init__(self, samples, sample_rate, channels, byte_width):
+        self.samples = str(base64.urlsafe_b64encode(samples), encoding='utf-8')
+        self.sample_rate = sample_rate
+        self.channels = channels
+        self.byte_width = byte_width
+
+    def to_json(self):
+        '''
+        类转json
+        '''
+        return json.dumps(self, default=lambda o: o.__dict__,
+            sort_keys=True)
+
+    def from_json(self, **entries):
+        '''
+        json转AsrtApiSpeechRequest
+        '''
+        self.__dict__.update(entries)
+
+    def __str__(self):
+        '''
+        AsrtApiSpeechRequest转为字符串
+        '''
+        return self.to_json()
+
+class AsrtApiLanguageRequest:
+    '''
+    ASRT语音识别基于HTTP协议的API接口请求类(声学模型)
+    '''
+    def __init__(self, sequence_pinyin):
+        self.sequence_pinyin = sequence_pinyin
+
+    def to_json(self):
+        '''
+        类转json
+        '''
+        return json.dumps(self, default=lambda o: o.__dict__,
+            sort_keys=True)
+
+    def from_json(self, **entries):
+        '''
+        json转AsrtApiLanguageRequest
+        '''
+        self.__dict__.update(entries)
+
+    def __str__(self):
+        '''
+        AsrtApiLanguageRequest转为字符串
+        '''
+        return self.to_json()
+
+class AsrtApiResponse:
+    '''
+    ASRT语音识别基于HTTP协议的API接口响应类
+    '''
+    def __init__(self, status_code=0, status_message='', result=''):
+        self.status_code = status_code
+        self.status_message = status_message
+        self.result = result
+    def to_json(self):
+        '''
+        类转json
+        '''
+        return json.dumps(self, default=lambda o: o.__dict__,
+            sort_keys=True)
+
+    def from_json(self, **entries):
+        '''
+        json转AsrtApiResponse
+        '''
+        self.__dict__.update(entries)
+
+    def __str__(self):
+        '''
+        AsrtApiResponse转为字符串
+        '''
+        return self.to_json()
+
+class WaveData:
+    '''
+    WAVE格式音频数据类
+    '''
+    def __init__(self, str_data, frame_rate, channels, byte_width) -> None:
+        self.str_data = str_data
+        self.sample_rate = frame_rate
+        self.channels = channels
+        self.byte_width = byte_width
+        self.filename = ''
+
+    def get_samples(self):
+        '''
+        str_data转short数组
+        '''
+        # 将声音文件数据转换为数组矩阵形式
+        wave_data = np.fromstring(self.str_data, dtype = np.short)
+        # 按照声道数将数组整形，单声道时候是一列数组，双声道时候是两列的矩阵
+        wave_data.shape = -1, self.channels
+        # 将矩阵转置
+        wave_data = wave_data.T
+        return wave_data
+
+    def set_filename(self, filename):
+        '''
+        记录该wave文件名
+        '''
+        self.filename = filename
+
+def read_wav_datas(filename):
+    '''
+    读取wave格式文件数据
+    '''
+    wav_file = wave.open(filename, 'rb')
+    num_frame = wav_file.getnframes()
+    str_data = wav_file.readframes(num_frame)
+    frame_rate = wav_file.getframerate()
+    channels = wav_file.getnchannels()
+    byte_width = wav_file.getsampwidth()
+    wav_file.close()
+    return WaveData(str_data, frame_rate, channels, byte_width)