Add files via upload

Author: JingShing

script/main.py

@@ -1,9 +1,20 @@
-from recording_voice import Recording_Helper
 from voice_detect import Voice_Detect_Helper
 
+def using_audio_to_talk(file_name = 'test.wav'):
+    from recording_voice import Recording_Helper
+    recorder = Recording_Helper(file_name=file_name)
+    voice_detecter = Voice_Detect_Helper(file_path=file_name)
+    recorder.recording_voice()
+    voice_detecter.detect_voice()
+    # you can get result from voice_detecter.result
+
+def using_detecter_to_talk(file_name = 'test.wav'):
+    voice_detecter = Voice_Detect_Helper(file_path=file_name)
+    voice_detecter.listener()
+
 if __name__ == '__main__':
-    recorder = Recording_Helper()
-    voice_detecter = Voice_Detect_Helper()
-    recorder()
-    voice_detecter()
-    # you can get result from voice_detecter.result
+    # method 1 using pyaudio to record
+    # using_audio_to_talk()
+
+    # method 2 using speech_recognition to record
+    using_detecter_to_talk()

script/recording_voice.py

@@ -2,7 +2,7 @@
 import wave
 
 class Recording_Helper:
-    def __init__(self, chunk = 1024, sample_format = pyaudio.paInt16, channels = 2, fs = 44100, seconds = 5, filename = "test.wav"):
+    def __init__(self, chunk = 1024, sample_format = pyaudio.paInt16, channels = 2, fs = 44100, seconds = 5, file_name = "test.wav"):
         # sample chunk size
         self.chunk = chunk
         # sample format: paFloat32, paInt32, paInt24, paInt16, paInt8, paUInt8, paCustomFormat
@@ -14,7 +14,7 @@ def __init__(self, chunk = 1024, sample_format = pyaudio.paInt16, channels = 2,
         # recording seconds
         self.seconds = seconds
         # recording file name
-        self.filename = filename
+        self.file_name = file_name
 
         self.p = pyaudio.PyAudio()
         # init pyaudio object
@@ -42,7 +42,7 @@ def recording_voice(self):
         print('stop recording...')
 
         # open voice file
-        wf = wave.open(self.filename, 'wb')
+        wf = wave.open(self.file_name, 'wb')
         # set channel
         wf.setnchannels(self.channels)
         # set format

script/voice_detect.py

@@ -7,11 +7,25 @@ def __init__(self, language='zh-TW', file_path="test.wav"):
         self.r = sr.Recognizer()
         self.result = None
     def detect_voice(self):
-        with sr.WavFile(self.file_path) as source:
+        with sr.AudioFile(self.file_path) as source:
+        # also you can use with sr.WavFile(self.file_path) as source:
+            
             # reduce ambient sound and noise
             # self.r.adjust_for_ambient_noise(source)
+
             # load wav
             audio = self.r.record(source)
+
+        try:
+            # using google service
+            self.result = self.r.recognize_google(audio,language=self.language)
+            print("Transcription: " + self.result)
+        except:
+            print("Could not understand audio")
+
+    def listener(self):
+        with sr.Microphone() as source:
+           audio = self.r.listen(source)
         try:
             # using google service
             self.result = self.r.recognize_google(audio,language=self.language)
@@ -20,4 +34,4 @@ def detect_voice(self):
             print("Could not understand audio")
 
     def __call__(self):
-        self.detect_voice()
+        self.listener()