Spaces:
Sleeping
Sleeping
File size: 2,302 Bytes
ea41881 11f2c2b ea41881 11f2c2b ea41881 11f2c2b ea41881 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 |
'''
Author: wuxulong19950206 [email protected]
Date: 2024-03-09 15:53:33
LastEditors: wuxulong19950206 [email protected]
LastEditTime: 2024-03-10 19:43:46
FilePath: \ASRT_SpeechRecognition\train_speech_model.py
Description: 这是默认设置,请设置`customMade`, 打开koroFileHeader查看配置 进行设置: https://github.com/OBKoro1/koro1FileHeader/wiki/%E9%85%8D%E7%BD%AE
'''
# !/usr/bin/env python3
# -*- coding: utf-8 -*-
#
# Copyright 2016-2099 Ailemon.net
#
# This file is part of ASRT Speech Recognition Tool.
#
# ASRT is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
# ASRT is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with ASRT. If not, see <https://www.gnu.org/licenses/>.
# ============================================================================
"""
@author: nl8590687
用于训练语音识别系统语音模型的程序
"""
import os
from tensorflow.keras.optimizers import Adam
from speech_model import ModelSpeech
from model_zoo.speech_model.keras_backend import SpeechModel251BN
from data_loader import DataLoader
from speech_features import SpecAugment
os.environ["CUDA_VISIBLE_DEVICES"] = "0"
AUDIO_LENGTH = 1600
AUDIO_FEATURE_LENGTH = 200
CHANNELS = 1
# 默认输出的拼音的表示大小是1428，即1427个拼音+1个空白块
OUTPUT_SIZE = 2884
sm251bn = SpeechModel251BN(
input_shape=(AUDIO_LENGTH, AUDIO_FEATURE_LENGTH, CHANNELS),
output_size=OUTPUT_SIZE
)
feat = SpecAugment()
train_data = DataLoader('train')
opt = Adam(learning_rate=0.0001, beta_1=0.9, beta_2=0.999, decay=0.0, epsilon=10e-8)
ms = ModelSpeech(sm251bn, feat, max_label_length=64)
ms.load_model("save_models\SpeechModel251bn\SpeechModel251bn_epoch17.model.h5")
ms.train_model(optimizer=opt, data_loader=train_data,
epochs=50, save_step=1, batch_size=1, last_epoch=0)
ms.save_model('save_models/' + sm251bn.get_model_name())
|