From 437da54bb801e73e9233fdeded12561518912f5f Mon Sep 17 00:00:00 2001
From: Apostolof <apotwohd@gmail.com>
Date: Mon, 7 Jan 2019 22:05:11 +0200
Subject: [PATCH] Init real time classification

---
 .../{test => }/accuracyWithoutFeature.py      |  6 +-
 classifier/pipeline.py                        |  2 +-
 classifier/streamClassify.py                  | 85 +++++++++++++++++++
 3 files changed, 88 insertions(+), 5 deletions(-)
 rename classifier/{test => }/accuracyWithoutFeature.py (74%)
 create mode 100644 classifier/streamClassify.py

diff --git a/classifier/test/accuracyWithoutFeature.py b/classifier/accuracyWithoutFeature.py
similarity index 74%
rename from classifier/test/accuracyWithoutFeature.py
rename to classifier/accuracyWithoutFeature.py
index 8d116f9..dd36340 100644
--- a/classifier/test/accuracyWithoutFeature.py
+++ b/classifier/accuracyWithoutFeature.py
@@ -1,14 +1,12 @@
 import numpy as np
 import pandas as pd
-from sys import path
-path.append('..')
 from feature_extraction.batch_feature_extractor import batchExtract
 from preprocessing.data_preprocessing import standardization
 from training.model_training import simpleTrain
 
-musicFeatures = batchExtract('../../dataset/music_wav/', '../feature_extraction/music_features/', 22050)
+musicFeatures = batchExtract('../dataset/music_wav/', '../feature_extraction/music_features/', 22050)
 musicFeatures = musicFeatures.assign(target=0)
-speechFeatures = batchExtract('../../dataset/speech_wav/', '../feature_extraction/speech_features/', 22050)
+speechFeatures = batchExtract('../dataset/speech_wav/', '../feature_extraction/speech_features/', 22050)
 speechFeatures = speechFeatures.assign(target=1)
 
 dataset = pd.concat([musicFeatures, speechFeatures])
diff --git a/classifier/pipeline.py b/classifier/pipeline.py
index 82150d7..79d02bd 100644
--- a/classifier/pipeline.py
+++ b/classifier/pipeline.py
@@ -14,7 +14,7 @@ dataset = pd.concat([musicFeatures, speechFeatures])
 target = dataset.pop('target').values
 
 dataset = standardization(dataset)
-# dataset = PCA(dataset)
+# _, dataset = PCA(dataset)
 
 print('Simple train accuracy achieved = ' + str(simpleTrain(dataset, target)))
 kFCrossValid(dataset, target, model = 'svm')
diff --git a/classifier/streamClassify.py b/classifier/streamClassify.py
new file mode 100644
index 0000000..699257d
--- /dev/null
+++ b/classifier/streamClassify.py
@@ -0,0 +1,85 @@
+import pyaudio
+import struct
+import math
+import pandas as pd
+
+from feature_extraction.feature_extractor import extractFeatures
+from preprocessing.data_preprocessing import createSingleFeaturesArray, standardization
+from training.model_training import kFCrossValid, simpleTrain
+
+FORMAT = pyaudio.paInt16
+SHORT_NORMALIZE = (1.0/32768.0)
+CHANNELS = 1
+RATE = 22050
+INPUT_BLOCK_TIME = int(6144 / RATE)
+INPUT_FRAMES_PER_BLOCK = 6144
+
+def classify(block, clf):
+	import numpy as np
+
+	audio_data = np.fromstring(block, np.int16)
+	audio_data = audio_data.astype(int)
+
+	values = extractFeatures(audio_data, 'tmp.json', RATE)
+	audioClass = clf.predict(values)
+
+	return audioClass
+
+class micListener(object):
+	def __init__(self):
+		self.pa = pyaudio.PyAudio()
+		self.stream = self.open_mic_stream()
+
+		print('Training...')
+		dataset, target, featureKeys = createSingleFeaturesArray(
+			'feature_extraction/music_features/',
+			'feature_extraction/speech_features/')
+
+		self.clf = kFCrossValid(dataset, target, model = 'rndForest')
+		print('Training done!')
+
+	def stop(self):
+		self.stream.close()
+
+	def find_input_device(self):
+		device_index = None
+		for i in range( self.pa.get_device_count() ):
+			devinfo = self.pa.get_device_info_by_index(i)
+			print( "Device %d: %s"%(i,devinfo["name"]) )
+
+			for keyword in ["mic","input"]:
+				if keyword in devinfo["name"].lower():
+					print( "Found an input: device %d - %s"%(i,devinfo["name"]) )
+					device_index = i
+					return device_index
+
+		if device_index == None:
+			print( "No preferred input found; using default input device." )
+
+		return device_index
+
+	def open_mic_stream( self ):
+		device_index = self.find_input_device()
+
+		stream = self.pa.open(format = FORMAT, channels = CHANNELS, rate = RATE,
+			input = True, input_device_index = device_index,
+			frames_per_buffer = INPUT_FRAMES_PER_BLOCK)
+
+		return stream
+
+	def listen(self):
+		try:
+			block = self.stream.read(INPUT_FRAMES_PER_BLOCK)
+		except IOError as e:
+			# dammit.
+			print('IOError!')
+			return
+
+		audioClass = classify(block, self.clf)
+		print(audioClass)
+
+if __name__ == "__main__":
+	micStr = micListener()
+
+	for i in range(300):
+		micStr.listen()
\ No newline at end of file