fix bugs and do some test

nl8590687 · May 8, 2018 · 08f7033 · 08f7033
1 parent bf489b9
commit 08f7033
Show file tree

Hide file tree

Showing 4 changed files with 21 additions and 10 deletions.
diff --git a/.gitignore b/.gitignore
@@ -11,3 +11,5 @@ dataset
 data_pinyin.txt
 
 testClient.py
+
+webapi/
diff --git a/SpeechModel22.py b/SpeechModel22.py
@@ -208,7 +208,7 @@ def TestModel(self, datapath='', str_dataset='dev', data_count = 32, out_report
 			if(out_report == True):
 				txt_obj = open('Test_Report_' + str_dataset + '_' + nowtime + '.txt', 'w', encoding='UTF-8') # 打开文件并读入
 
-			txt = ''
+
 			for i in range(data_count):
 				data_input, data_labels = data.GetData((ran_num + i) % num_data)  # 从随机数开始连续向后取一定数量数据
 				pre = self.Predict(data_input, data_input.shape[0] // 4)
@@ -221,16 +221,20 @@ def TestModel(self, datapath='', str_dataset='dev', data_count = 32, out_report
 				else: # 否则肯定是增加了一堆乱七八糟的奇奇怪怪的字
 					word_error_num += words_n # 就直接加句子本来的总字数就好了
 
+				if(i % 10 == 0):
+					print('测试进度：',i,'/',data_count)
+
+				txt = ''
 				if(out_report == True):
 					txt += str(i) + '\n'
 					txt += 'True:\t' + str(data_labels) + '\n'
 					txt += 'Pred:\t' + str(pre) + '\n'
 					txt += '\n'
-
+				txt_obj.write(txt)
 
 			print('*[测试结果] 语音识别 ' + str_dataset + ' 集语音单字错误率：', word_error_num / words_num * 100, '%')
 			if(out_report == True):
-				txt += '*[测试结果] 语音识别 ' + str_dataset + ' 集语音单字错误率： ' + str(word_error_num / words_num * 100) + ' %'
+				txt = '*[测试结果] 语音识别 ' + str_dataset + ' 集语音单字错误率： ' + str(word_error_num / words_num * 100) + ' %'
 				txt_obj.write(txt)
 				txt_obj.close()
 
@@ -381,9 +385,9 @@ def model(self):
 
 	ms = ModelSpeech(datapath)
 
-	#ms.LoadModel(modelpath + 'm22_2\\speech_model22_e_0_step_257000.model')
+	#ms.LoadModel(modelpath + 'm22_2\\1\\speech_model22_e_0_step_159000.model')
 	ms.TrainModel(datapath, epoch = 50, batch_size = 4, save_step = 500)
-	#ms.TestModel(datapath, str_dataset='test', data_count = 64, out_report = True)
+	#ms.TestModel(datapath, str_dataset='test', data_count = 128, out_report = True)
 	#r = ms.RecognizeSpeech_FromFile('E:\\语音数据集\\ST-CMDS-20170001_1-OS\\20170001P00241I0053.wav')
 	#r = ms.RecognizeSpeech_FromFile('E:\\语音数据集\\ST-CMDS-20170001_1-OS\\20170001P00020I0087.wav')
 	#r = ms.RecognizeSpeech_FromFile('E:\\语音数据集\\wav\\train\\A11\\A11_167.WAV')

diff --git a/asrserver.py b/asrserver.py
@@ -4,7 +4,6 @@
 @author: nl8590687
 语音识别API的HTTP服务器程序
 
-尚未完成
 """
 import http.server
 import urllib
@@ -45,6 +44,7 @@ def do_POST(self):
 		token = ''
 		fs = 0
 		wavs = []
+		#type = 'wavfilebytes' # wavfilebytes or python-list
 
 		for line in datas_split:
 			[key, value]=line.split('=')
@@ -54,10 +54,15 @@ def do_POST(self):
 				fs = int(value)
 			elif('token' == key ):
 				token = value
+			#elif('type' == key):
+			#	type = value
 			else:
 				print(key, value)
 
+		#if('python-list' == type):
 		r = self.recognize([wavs], fs)
+		#else:
+		#	r = self.recognize_from_file('')
 
 		if(token == 'qwertasd'):
 			#buf = '成功\n'+'wavs:\n'+str(wavs)+'\nfs:\n'+str(fs)
@@ -74,7 +79,7 @@ def do_POST(self):
 
 
 		#buf = '<!DOCTYPE HTML> \n<html> \n<head>\n<title>Post page</title>\n</head> \n<body>Post Data:%s  <br />Path:%s\n</body>  \n</html>'%(datas,self.path)  
-		buf = bytes(buf],encoding="utf-8")
+		buf = bytes(buf,encoding="utf-8")
 		self.wfile.write(buf)  
 
 	def recognize(self, wavs, fs):
@@ -92,6 +97,9 @@ def recognize(self, wavs, fs):
 		return r
 		pass
 
+	def recognize_from_file(self, filename):
+		pass
+
 def start_server(ip, port):  
 	http_server = http.server.HTTPServer((ip, int(port)), TestHTTPHandle)  
 	print('服务器已开启')

diff --git a/test.py b/test.py
@@ -33,9 +33,6 @@
 
 #ms.TestModel(datapath, str_dataset='test', data_count = 64, out_report = True)
 r = ms.RecognizeSpeech_FromFile('E:\语音数据集\ST-CMDS-20170001_1-OS\\20170001P00241I0052.wav')
-#r = ms.RecognizeSpeech_FromFile('E:\\VS2015解决方案\\wav文件读写样例\\wav文件读写样例\\bin\\Debug\\1.wav')
-#r = ms.RecognizeSpeech_FromFile('/home/nl/01.wav')
-#r = ms.RecognizeSpeech_FromFile('C:\\Users\\nl\\Desktop\\01.wav')
 #r = ms.RecognizeSpeech_FromFile('E:\语音数据集\ST-CMDS-20170001_1-OS\\20170001P00241I0053.wav')
 #r = ms.RecognizeSpeech_FromFile('E:\\语音数据集\\ST-CMDS-20170001_1-OS\\20170001P00020I0087.wav')
 #r = ms.RecognizeSpeech_FromFile('E:\\语音数据集\\wav\\train\\A11\\A11_167.WAV')
Original file line number	Diff line number	Diff line change
Expand Up		@@ -11,3 +11,5 @@ dataset
		data_pinyin.txt

		testClient.py

		webapi/