py/api.py

@@ -99,9 +99,16 @@ class API:
 
         @self.app.route('/interstellar_ai/api/voice_recognition', methods=['POST'])
         def voice_recognition():
+            print(request.args)
+            recog_type = request.form.get('type')
+            print(recog_type)
             audio = request.files.get('audio')
-            text = self.voice.recognition(audio)
+            option = request.form.get('option')
+            if recog_type == "basic":
+                text = self.voice.basic_recognition(audio, option)
                 return jsonify({'status': 200, 'response': text})
+            else:
+                return jsonify({'status': 401, 'response': "Invalid type"})
 
         @self.app.route('/interstellar_ai/api/weather', methods=['POST'])
         def get_weather():

py/requirements.txt

@@ -14,4 +14,3 @@ pycouchdb
 pyttsx3
 pip-licenses
 openai-whisper
-pydub

py/voice.py

@@ -1,10 +1,24 @@
-import whisper
+import speech_recognition as sr
 
 
 class VoiceRecognition:
     @staticmethod
-    def recognition(audio):
-        model = whisper.load_model("base")
-        result = model.transcribe(audio)
-        print(result["text"])
-        return result["text"]
+    def basic_recognition(audio, option):
+        print(type(audio))
+        print("preparing")
+        r = sr.Recognizer()
+        if option == "online":
+            with sr.AudioFile(audio) as source:
+                print("online")
+                text = r.recognize_google_cloud(source)
+                print("recognized as: " + text)
+                return text
+        elif option == "offline":
+            with sr.AudioFile(audio) as source:
+                print("offline")
+                text = r.recognize_sphinx(source)
+                print("recognized as: " + text)
+                return text
+
+        print("nothing")
+        return False