def init_model():
global x, y
# set log level to debug
tf.sg_verbosity(10)
#
# hyper parameters
#
batch_size = 1 # batch size
#
# inputs
#
# vocabulary size
voca_size = data.voca_size
# print(voca_size)
# mfcc feature of audio
x = tf.placeholder(dtype=tf.sg_floatx, shape=(batch_size, None, 20))
# sequence length except zero-padding
seq_len = tf.not_equal(x.sg_sum(axis=2), 0.).sg_int().sg_sum(axis=1)
# encode audio feature
logit = get_logit(x, voca_size=voca_size)
# ctc decoding
decoded, _ = tf.nn.ctc_beam_search_decoder(
logit.sg_transpose(perm=[1, 0, 2]), seq_len, merge_repeated=False)
# to dense tensor
y = tf.sparse_to_dense(decoded[0].indices, decoded[0].dense_shape,
decoded[0].values) + 1
def run(self, datafile):
# ctc decoding
decoded, _ = tf.nn.ctc_beam_search_decoder(
self.logit.sg_transpose(perm=[1, 0, 2]),
self.seq_len,
merge_repeated=False)
# to dense tensor
y = tf.sparse_to_dense(decoded[0].indices, decoded[0].dense_shape,
decoded[0].values) + 1
# load wave file
wav, _ = librosa.load(datafile, mono=True, sr=16000)
# get mfcc feature
mfcc = np.transpose(
np.expand_dims(librosa.feature.mfcc(wav, 16000), axis=0),
[0, 2, 1])
# run session
label = self.session.run(y, feed_dict={self.x: mfcc})
# return string
return sttwdata.index_as_string(label)
from model import *
import data
batch_size = 1 # batch size
voca_size = data.voca_size
x = tf.placeholder(dtype=tf.sg_floatx, shape=(batch_size, None, 20))
# sequence length except zero-padding
seq_len = tf.not_equal(x.sg_sum(axis=2), 0.).sg_int().sg_sum(axis=1)
# encode audio feature
logit = get_logit(x, voca_size)
# ctc decoding
decoded, _ = tf.nn.ctc_beam_search_decoder(logit.sg_transpose(perm=[1, 0, 2]), seq_len, merge_repeated=False)
# to dense tensor
y = tf.add(tf.sparse_to_dense(decoded[0].indices, decoded[0].dense_shape, decoded[0].values), 1, name="output")
with tf.Session() as sess:
tf.sg_init(sess)
saver = tf.train.Saver()
saver.restore(sess, tf.train.latest_checkpoint('asset/train'))
graph = tf.get_default_graph()
input_graph_def = graph.as_graph_def()
with tf.Session() as sess:
tf.sg_init(sess)
saver = tf.train.Saver()
saver.restore(sess, tf.train.latest_checkpoint('asset/train'))
# Output model's graph details for reference.
tf.train.write_graph(sess.graph_def, '/root/speech-to-text-wavenet/asset/train', 'graph.txt', as_text=True)
# menginput mfcc feature pada file audio
x = tf.placeholder(dtype=tf.sg_floatx, shape=(batch_size, None, 20))
# panjang sequence kecuali zero-padding
seq_len = tf.not_equal(x.sg_sum(axis=2), 0.).sg_int().sg_sum(axis=1)
# encode audio feature
logit = get_logit(x, voca_size=voca_size)
# ctc decoding
decoded, _ = tf.nn.ctc_beam_search_decoder(logit.sg_transpose(perm=[1, 0, 2]),
seq_len,
merge_repeated=False)
# to dense tensor
y = tf.sparse_to_dense(decoded[0].indices, decoded[0].dense_shape,
decoded[0].values) + 1
# regcognize audio file
# perintah untuk menginput path file audio
tf.sg_arg_def(file=('', 'speech wave file to recognize.'))
# load audio file
file = sys.argv[1]
wav, sr = librosa.load(file, mono=True, sr=16000)
# mendapatkan mfcc feature
mfcc = np.transpose(np.expand_dims(librosa.feature.mfcc(wav, 16000), axis=0),
[0, 2, 1])
# run network