Python ModelSpeech Exemples

Exemple #1

datapath = ''
modelpath = 'model_speech'

system_type = plat.system()  # 由于不同的系统的文件路径表示不一样，需要进行判断
if (system_type == 'Windows'):
    datapath = 'E:\\语音数据集'
    modelpath = modelpath + '\\'
elif (system_type == 'Linux'):
    datapath = 'dataset'
    modelpath = modelpath + '/'
else:
    print('*[Message] Unknown System\n')
    datapath = 'dataset'
    modelpath = modelpath + '/'

ms = ModelSpeech(datapath)

ms.LoadModel(modelpath + 'm_dfcnn/speech_model_dfcnn_e_0_step_' + EPOCH +
             '000.model')
#ms.LoadModel(modelpath + 'm_DFCNN/speech_model_DFCNN_e_0_step_8000.model')
#ms.LoadModel(modelpath + 'm26/speech_model26_e_0_step_122500.model')

#ms.TestModel(datapath, str_dataset='test', data_count = 64, out_report = True)
#r = ms.RecognizeSpeech_FromFile('myspeech.wav')
#r = ms.RecognizeSpeech_FromFile('/home/speech.AI/github/DFCNN/dataset/data_thchs30/test/D11_750.wav')
#aishell
r = ms.RecognizeSpeech_FromFile(
    '/home/speech.AI/github/DFCNN/dataset/data_aishell/wav/test/S0764/BAC009S0764W0122.wav'
)
#r = ms.RecognizeSpeech_FromFile('/home/speech.AI/github/DFCNN/dataset/data_aishell/wav/test/S0764/BAC009S0764W0124.wav')
#r = ms.RecognizeSpeech_FromFile('/home/speech.AI/github/DFCNN/dataset/data_aishell/wav/test/S0764/BAC009S0764W0179.wav')

Exemple #2

#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
语音识别API的HTTP服务器程序

"""
import http.server
import urllib
import keras
from SpeechModel_DFCNN import ModelSpeech
from LanguageModel import ModelLanguage

datapath = 'data/'
modelpath = 'model_speech/'
ms = ModelSpeech(datapath)
ms.LoadModel(modelpath + 'm_DFCNN/speech_model_DFCNN_e_0_step_410000.model')

ml = ModelLanguage('model_language')
ml.LoadModel()


class TestHTTPHandle(http.server.BaseHTTPRequestHandler):
    def setup(self):
        self.request.settimeout(10)
        http.server.BaseHTTPRequestHandler.setup(self)

    def _set_response(self):
        self.send_response(200)
        self.send_header('Content-type', 'text/html')
        self.end_headers()

Exemple #3

datapath = ''
modelpath = 'model_speech'

f = open('step_dfcnn.txt', 'r')
flist = f.readlines()
f.close()
fstr = "".join(flist)
base_count = fstr.split('_')[-1]

if (not os.path.exists(modelpath)):  # 判断保存模型的目录是否存在
    os.makedirs(modelpath)  # 如果不存在，就新建一个，避免之后保存模型的时候炸掉

system_type = plat.system()  # 由于不同的系统的文件路径表示不一样，需要进行判断
if (system_type == 'Windows'):
    datapath = 'E:\\语音数据集'
    modelpath = modelpath + '\\'
elif (system_type == 'Linux'):
    datapath = 'dataset'
    modelpath = modelpath + '/'
else:
    print('*[Message] Unknown System\n')
    datapath = 'dataset'
    modelpath = modelpath + '/'

ms = ModelSpeech(datapath)

#ms.LoadModel(modelpath + 'm_DFCNN/speech_model_DFCNN_e_0_step_84000.model')
#ms.LoadModel(modelpath + 'speech_model_dfcnn_e_0_step_' + base_count +'.model')
ms.TrainModel(datapath, epoch=50, batch_size=16, save_step=1000)

Exemple #4

modelpath = 'model_speech'

f = open('step_dfcnn.txt','r')
flist = f.readlines()
f.close()
fstr = "".join(flist)
base_count=fstr.split('_')[-1]

if(not os.path.exists(modelpath)): # 判断保存模型的目录是否存在
	os.makedirs(modelpath) # 如果不存在，就新建一个，避免之后保存模型的时候炸掉

system_type = plat.system() # 由于不同的系统的文件路径表示不一样，需要进行判断
if(system_type == 'Windows'):
	datapath = 'E:\\语音数据集'
	modelpath = modelpath + '\\'
elif(system_type == 'Linux'):
	datapath = 'dataset'
	modelpath = modelpath + '/'
else:
	print('*[Message] Unknown System\n')
	datapath = 'dataset'
	modelpath = modelpath + '/'

ms = ModelSpeech(datapath)

#ms.LoadModel(modelpath + 'm_DFCNN/speech_model_DFCNN_e_0_step_84000.model')
ms.LoadModel(modelpath + 'm_dfcnn/speech_model_dfcnn_e_0_step_' + base_count +'.model')
ms.TrainModel(datapath, epoch = 50, batch_size = 64, save_step = 1000)

Exemple #5

config = tf.ConfigProto()
config.gpu_options.per_process_gpu_memory_fraction = 0.9
#config.gpu_options.allow_growth=True   #不全部占满显存, 按需分配
set_session(tf.Session(config=config))
'''
os.environ["CUDA_DEVICE_ORDER"] = "PCI_BUS_ID"
os.environ["CUDA_VISIBLE_DEVICES"] = "0, 1"
#进行配置，使用70%的GPU
config = tf.ConfigProto(allow_soft_placement=True)
config.gpu_options.per_process_gpu_memory_fraction = 0.95
#config.gpu_options.allow_growth=True   #不全部占满显存, 按需分配
set_session(tf.Session(config=config))

datapath = ''
modelpath = 'model_speech/'

if (not os.path.exists(modelpath)):  # 判断保存模型的目录是否存在
    os.makedirs(modelpath)  # 如果不存在，就新建一个，避免之后保存模型的时候炸掉

ms = ModelSpeech(datapath)

ms.LoadModel(modelpath + 'm_dfcnn/speech_model_dfcnn_e_0_step_64000.model')
#ms.LoadModel(modelpath + 'm_DFCNN/speech_model_DFCNN_e_0_step_410000.model')
ms.TestModel(datapath, str_dataset='test', data_count=64, out_report=True)

#r = ms.RecognizeSpeech_FromFile('E:\\语音数据集\\ST-CMDS-20170001_1-OS\\20170001P00241I0053.wav')
#r = ms.RecognizeSpeech_FromFile('E:\\语音数据集\\ST-CMDS-20170001_1-OS\\20170001P00020I0087.wav')
#r = ms.RecognizeSpeech_FromFile('E:\\语音数据集\\wav\\train\\A11\\A11_167.WAV')
#r = ms.RecognizeSpeech_FromFile('E:\\语音数据集\\wav\\test\\D4\\D4_750.wav')
#print('*[提示] 语音识别结果：\n',r)

Exemple #6

Fichier : test.py Projet : ramosmy/DFCNN

from keras.backend.tensorflow_backend import set_session
from SpeechModel_DFCNN import ModelSpeech
from LanguageModel import ModelLanguage

os.environ["CUDA_DEVICE_ORDER"] = "PCI_BUS_ID"
os.environ["CUDA_VISIBLE_DEVICES"] = "0, 1"
#进行配置，使用70%的GPU
config = tf.ConfigProto(allow_soft_placement=True)
config.gpu_options.per_process_gpu_memory_fraction = 0.9
#config.gpu_options.allow_growth=True   #不全部占满显存, 按需分配
set_session(tf.Session(config=config))

datapath = ''
modelpath = 'model_speech/'

ms = ModelSpeech(datapath)

ms.LoadModel(modelpath + 'm_dfcnn/speech_model_dfcnn_e_0_step_64000.model')
#ms.LoadModel(modelpath + 'm_DFCNN/speech_model_DFCNN_e_0_step_410000.model')
#ms.LoadModel(modelpath + 'm26/speech_model26_e_0_step_122500.model')

#ms.TestModel(datapath, str_dataset='test', data_count = 64, out_report = True)
#r = ms.RecognizeSpeech_FromFile('/home/speech.AI/github/DFCNN/dataset/data_thchs30/test/D11_750.wav')
r = ms.RecognizeSpeech_FromFile(
    '/home/speech.AI/github/DFCNN/dataset/data_thchs30/train/A33_100.wav')
print('*[提示] 语音识别结果：\n', r)

ml = ModelLanguage('model_language')
ml.LoadModel()

#str_pinyin = ['zhe4','zhen1','shi4','ji2', 'hao3','de5']

Exemple #7

Fichier : test_mspeech.py Projet : woshidafengzi/DFCNN-master

if(not os.path.exists(modelpath)): # 判断保存模型的目录是否存在
	os.makedirs(modelpath) # 如果不存在，就新建一个，避免之后保存模型的时候炸掉

system_type = plat.system() # 由于不同的系统的文件路径表示不一样，需要进行判断
if(system_type == 'Windows'):
	datapath = 'E:\\语音数据集'
	modelpath = modelpath + '\\'
elif(system_type == 'Linux'):
	datapath = 'dataset'
	modelpath = modelpath + '/'
else:
	print('*[Message] Unknown System\n')
	datapath = 'dataset'
	modelpath = modelpath + '/'

ms = ModelSpeech(datapath)

ms.LoadModel(modelpath + 'm_dfcnn/speech_model_dfcnn_e_0_step_'+iters_num+'000.model')
#ms.LoadModel(modelpath + 'm_DFCNN/speech_model_DFCNN_e_0_step_84000.model')
ms.TestModel(datapath, str_dataset='test', data_count = 64, out_report = True)

r = ms.RecognizeSpeech_FromFile('dataset/data_aishell/wav/dev/S0733/BAC009S0733W0234.wav')
#r = ms.RecognizeSpeech_FromFile('E:\\语音数据集\\ST-CMDS-20170001_1-OS\\20170001P00020I0087.wav')
#r = ms.RecognizeSpeech_FromFile('E:\\语音数据集\\wav\\train\\A11\\A11_167.WAV')
#r = ms.RecognizeSpeech_FromFile('E:\\语音数据集\\wav\\test\\D4\\D4_750.wav')
print('*[提示] 语音识别结果：\n',r)