Speech recognition

scripts/recognition/audio.py

@@ -1,18 +1,48 @@
-from .emotion import Emotion
-from .phrase import Phrase
-
-def emotionFromAudio(audio) -> Emotion:
-    """
-    Extracts the emotion from a voice.
-    """
-    emotion = Emotion.NEUTRAL
-    # TODO: implement this
-    return emotion
-
-def phraseFromAudio(audio) -> Phrase:
-    """
-    Extracts the phrase from a voice.
-    """
-    phrase = Phrase.UNKNOWN
-    # TODO: implement this
-    return phrase
+# from .emotion import Emotion
+# from .phrase import Phrase
+
+from vosk import Model, KaldiRecognizer
+
+import pyaudio
+import json
+
+import signal
+import sys
+
+
+class Audio():
+
+    def __init__(self):
+        self.model = Model("../../data/vosk-model-small-en-us-0.15") 
+        self.recognizer = KaldiRecognizer(self.model,  16000)
+        self.stream = pyaudio.PyAudio().open(format= pyaudio.paInt16, channels=1, rate= 16000, input=True, frames_per_buffer=8192)
+        self.stream.start_stream()
+
+    # def emotionFromAudio(self) -> Emotion:
+    #     """
+    #     Extracts the emotion from a voice.
+    #     """
+    #     emotion = Emotion.NEUTRAL
+    #     # TODO: implement this
+    #     return emotion
+
+    def phraseFromAudio(self):
+        """
+        Extracts the phrase from the audio.
+        """
+        phrase = ""
+        while True:
+            try:
+                data = self.stream.read(4096)
+                if self.recognizer.AcceptWaveform(data):
+                    res = json.loads(self.recognizer.Result())
+                    phrase = res["text"]
+                    print(phrase)
+            except KeyboardInterrupt:
+                break
+
+        return phrase
+
+
+audio = Audio()
+audio.phraseFromAudio()

scripts/recognition/nlp.py

@@ -0,0 +1,3 @@
+import spacy
+
+#TODO: Naïve Bayes Classifier

scripts/recognition/phrase.py

@@ -6,3 +6,15 @@ class Phrase(Enum):
     HI = 1
     BYE = 2
     LOOK_AT_ME = 3
+
+    def __init__(self, phrase) -> None:
+        self.phrase = phrase
+
+    def setPhrase(self, phrase):
+        self.phrase = self
+
+    def getPhrase(self):
+        return self.phrase
+
+    def emotionFromPhrase(self):
+        # TODO: implement this

scripts/setup.py

@@ -1,12 +1,24 @@
 from core import Parser
+from zipfile import ZipFile
 
 import os
 import requests
 
+import sys
+import subprocess #https://docs.python.org/3/library/subprocess.html
+
+
 # Files to download into the data directory.
 FILES = [
     "https://raw.githubusercontent.com/kipr/opencv/master/data/haarcascades/haarcascade_frontalface_default.xml",
     "https://raw.githubusercontent.com/kipr/opencv/master/data/haarcascades/haarcascade_eye_tree_eyeglasses.xml",
+    "https://alphacephei.com/vosk/models/vosk-model-small-en-us-0.15.zip"
+]
+
+PACKAGES = [
+    "vosk", 
+    "spacy", 
+    "PyAudio"
 ]
 
 def setup(dataDir):
@@ -28,6 +40,28 @@ def setup(dataDir):
         filename = os.path.basename(file)
         path = os.path.join(dataDir, filename)
         open(path, "wb").write(response.content)
+
+    # Loading the temp.zip and creating a zip object
+    with ZipFile("../data/vosk-model-small-en-us-0.15.zip", 'r') as zObject:
+
+        # Extracting the zip
+        zObject.extractall(path="../data")
+
+    # Delete the zip file
+    os.remove("../data/vosk-model-small-en-us-0.15.zip")
+
+
+def install_packages():
+    """Install the required python libraries."""
+
+    for package in PACKAGES:
+        print("Downloading python library {}...".format(package))
+        try:
+            subprocess.check_call([sys.executable, "-m", "pip", "install", package], stdout=subprocess.DEVNULL)
+        except Exception as e:
+            print("Failed to download library {}".format(package))
+            print("Exception {}".format(e))
+
 
 if __name__ == "__main__":
     # Default data directory is `../data` relative to this file.
@@ -42,3 +76,4 @@ def setup(dataDir):
         print(parser.help())
     else:
         setup(options["data"])
+        install_packages()