Deployed e6c4a70 with MkDocs version: 1.6.1

2019/imgs/Leo/xml/Untitled Diagram.drawio

@@ -0,0 +1 @@
+<mxfile host="www.draw.io" modified="2019-12-05T18:28:28.327Z" agent="Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/73.0.3683.103 Safari/537.36" etag="4B4RHLqeErGk9YAUnP_C" version="12.3.7" type="device" pages="1"><diagram id="xq_sSW7vf1Avch4HxVfm" name="Page-1">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</diagram></mxfile>

2019/imgs/Leo/xml/Untitled Diagram.xml

@@ -0,0 +1 @@
+<mxfile host="www.draw.io" modified="2019-12-05T16:51:33.783Z" agent="Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/73.0.3683.103 Safari/537.36" etag="QeljNpgT-1b_vimL3-BH" version="12.3.7" type="device" pages="1"><diagram id="WPLDtqaatGJPYjKMvfv_" name="Page-1">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</diagram></mxfile>

2019/imgs/Leo/xml/Untitled Diagram2.xml

@@ -0,0 +1 @@
+<mxfile host="www.draw.io" modified="2019-12-05T16:52:05.234Z" agent="Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/73.0.3683.103 Safari/537.36" etag="gupGxY8cYijm6JhvRd67" version="12.3.7" type="device" pages="1"><diagram id="WPLDtqaatGJPYjKMvfv_" name="Page-1">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</diagram></mxfile>

2019/src/Gabriel-TensorFlow/audio_stream.py

@@ -0,0 +1,182 @@
+from __future__ import absolute_import
+from __future__ import division
+from __future__ import print_function
+
+import argparse
+import sys
+import pyaudio
+import numpy as np
+import audioop
+import wave
+import tensorflow as tf
+import time
+
+FLAGS = None
+IN_FILE = 'in.wav' # name of input file
+CHUNK = 4096 # number of data points to read at a time
+RATE = 16000 # time resolution of the recording device (Hz)
+CHANNELS = 1 # number of channels
+
+FORMAT = pyaudio.paInt16 # audio format from pyaudio
+p=pyaudio.PyAudio() # start the PyAudio class
+devinfo = p.get_device_info_by_index(0) # get the first recorder device
+
+# stream from pyaudio
+stream=p.open(format=FORMAT, channels=CHANNELS, rate=RATE, input=True,
+              frames_per_buffer=CHUNK)
+
+
+def load_graph(filename):
+    """Unpersists graph from file as default graph."""
+    with tf.io.gfile.GFile(filename, 'rb') as f:
+        graph_def = tf.compat.v1.GraphDef()
+        graph_def.ParseFromString(f.read())
+        tf.import_graph_def(graph_def, name='')
+
+
+def load_labels(filename):
+    """Read in labels, one label per line."""
+    return [line.rstrip() for line in tf.io.gfile.GFile(filename)]
+
+
+def run_graph(wav_data, labels, input_layer_name, output_layer_name,
+              num_top_predictions):
+  """Runs the audio data through the graph and prints predictions."""
+  with tf.compat.v1.Session() as sess:
+    # Feed the audio data as input to the graph.
+    #   predictions  will contain a two-dimensional array, where one
+    #   dimension represents the input image count, and the other has
+    #   predictions per class
+    softmax_tensor = sess.graph.get_tensor_by_name(output_layer_name)
+    predictions, = sess.run(softmax_tensor, {input_layer_name: wav_data})
+
+    # Sort to show labels in order of confidence
+    top_k = predictions.argsort()[-num_top_predictions:][::-1]
+    for node_id in top_k:
+        human_string = labels[node_id]
+        score = predictions[node_id]
+        print('%s (score = %.5f)' % (human_string, score))
+
+    return 0
+
+
+def label_wav(wav, labels, graph, input_name, output_name, how_many_labels):
+    """Loads the model and labels, and runs the inference to print predictions."""
+    if not wav or not tf.io.gfile.exists(wav):
+        tf.compat.v1.logging.fatal('Audio file does not exist %s', wav)
+
+    if not labels or not tf.io.gfile.exists(labels):
+        tf.compat.v1.logging.fatal('Labels file does not exist %s', labels)
+
+    if not graph or not tf.io.gfile.exists(graph):
+        tf.compat.v1.logging.fatal('Graph file does not exist %s', graph)
+
+    labels_list = load_labels(labels)
+
+    # load graph, which is stored in the default session
+    load_graph(graph)
+
+    with open(wav, 'rb') as wav_file:
+        wav_data = wav_file.read()
+
+    run_graph(wav_data, labels_list, input_name, output_name, how_many_labels)
+
+
+def prepare(graph, labels):
+    """Loads data labels and tensor graphs"""
+    labels_list = load_labels(labels)
+    load_graph(graph)
+    return labels_list
+
+def main(_):
+
+    # initialize variables and prepare graph
+    labels = prepare(FLAGS.graph, FLAGS.labels)
+    recording = False
+    frames = []
+    time_from_previous = 0
+
+    with tf.compat.v1.Session() as sess:
+
+        # loads softmax tensor
+        softmax_tensor = sess.graph.get_tensor_by_name(FLAGS.output_name)
+
+        while True:
+            # transform data into a numpy array
+            data = np.fromstring(stream.read(CHUNK), dtype=np.int16)
+
+            # get audio rms
+            rms = audioop.rms(data, 2)
+
+            # if audio rms reaches 900 or more set recording for true
+            # and start appending the data into the frames array.
+
+            # this means someone is talking 
+            if rms > 900: 
+                if recording == False:
+                    recording = True
+                frames.append(data)
+            else:
+                if recording == True:
+                    time_from_previous = time.time()
+
+                    # get data from stream for the next 0.5 seconds
+                    # after the volume 
+                    while time.time() - time_from_previous < 0.5:
+                        data = np.fromstring(stream.read(CHUNK),dtype=np.int16)
+                        frames.append(data)
+
+                    # write frames inside wav file
+                    _file = wave.open("out.wav","wb")
+                    _file.setnchannels(CHANNELS)
+                    _file.setsampwidth(p.get_sample_size(FORMAT))
+                    _file.setframerate(RATE)
+                    _file.writeframes(b''.join(frames))
+                    _file.close()
+
+                    # clear frames array since the data was
+                    # written inside wav file
+                    frames = []
+
+                    # read wav file to get the input data for
+                    # our neural network
+                    with open('out.wav', 'rb') as wav_file:
+                        wav_data = wav_file.read()
+
+                    # this is where the model predicts based on the input data
+                    predictions, = sess.run(softmax_tensor, {FLAGS.input_name: wav_data})
+
+		            # Sort to show labels in order of confidence
+                    top_k = predictions.argsort()[-FLAGS.how_many_labels:][::-1]
+                    for node_id in top_k:
+                        human_string = labels[node_id]
+                        score = predictions[node_id]
+                        print('%s (score = %.5f)' % (human_string, score))
+                recording = False
+
+if __name__ == '__main__':
+    parser = argparse.ArgumentParser()
+    parser.add_argument(
+        '--wav', type=str, default='', help='Audio file to be identified.')
+    parser.add_argument(
+        '--graph', type=str, default='', help='Model to use for identification.')
+    parser.add_argument(
+        '--labels', type=str, default='', help='Path to file containing labels.')
+    parser.add_argument(
+        '--input_name',
+        type=str,
+        default='wav_data:0',
+        help='Name of WAVE data input node in model.')
+    parser.add_argument(
+        '--output_name',
+        type=str,
+        default='labels_softmax:0',
+        help='Name of node outputting a prediction in the model.')
+    parser.add_argument(
+        '--how_many_labels',
+        type=int,
+        default=1,
+        help='Number of results to show.')
+
+    FLAGS, unparsed = parser.parse_known_args()
+    tf.compat.v1.app.run(main=main, argv=[sys.argv[0]] + unparsed)