46 Commits

Author	SHA1	Message	Date
liuhaozhe6788	2d63164845	new commits	2023-09-09 13:41:55 +08:00
liuhaozhe6788	74b4af24ed	add demo results	2023-07-23 15:52:41 +08:00
liuhaozhe6788	635148960c	new commits	2023-07-17 19:19:33 +08:00
liuhaozhe6788	ee5f4fb3da	skip denoising	2023-06-27 15:01:01 +08:00
liuhaozhe6788	3c447437b0	update toolbox	2023-06-25 17:28:39 +08:00
liuhaozhe6788	e7a3fd0a74	new commits	2023-06-25 10:49:22 +08:00
liuhaozhe6788	d55e88172f	new commits	2023-06-23 10:20:11 +08:00
liuhaozhe6788	000c3ad71f	new commits	2023-06-19 19:44:56 +08:00
liuhaozhe6788	bc7d67ddbf	new commits	2023-06-19 10:42:03 +08:00
liuhaozhe6788	38a2b49b61	new commits	2023-06-15 17:17:57 +08:00
liuhaozhe6788	c2a4fb9408	new commits	2023-06-15 16:03:48 +08:00
liuhaozhe6788	9034de82b6	bug fix for synthesizer training pipeline	2023-06-15 10:30:06 +08:00
liuhaozhe6788	b1dc8c8296	rename encoder	2023-06-09 10:47:49 +08:00
liuhaozhe6788	00be83f97e	new commits	2023-06-08 15:30:32 +08:00
liuhaozhe6788	ba7b119073	new commits	2023-06-08 15:21:34 +08:00
liuhaozhe6788	82176597ca	new commits	2023-06-03 19:52:49 +08:00
liuhaozhe6788	a1ae4132f1	new commits	2023-06-03 19:26:14 +08:00
liuhaozhe6788	b734f2c41c	调整tacotron停止符的判别参数，处理GPU预测显存占用大的问题	2023-04-18 17:43:58 +08:00
liuhaozhe6788	4cc64326e2	修复bug	2023-04-17 18:30:08 +08:00
liuhaozhe6788	d0409b8c07	修复bug并改善合成器预测终止判别参数	2023-04-17 18:27:42 +08:00
liuhaozhe6788	b77e01b96d	给TTS模块增加VCTK数据继续训练模型	2023-03-29 16:02:41 +08:00
liuhaozhe6788	b5296a9a36	remove temp files	2023-03-09 16:17:03 +08:00
liuhaozhe6788	df983336ac	解决单个字母发音和声音变速的bug，并改善guy的发音	2023-03-09 15:53:34 +08:00
liuhaozhe6788	3fcde2e668	只将最多含有3个字符的大写单词转为字符读音并识别复数形式的缩略词转为字符读音	2023-03-07 15:47:02 +08:00
liuhaozhe6788	4bec275e1a	转换更多ascii码符号，并对含有数字的缩写进行处理	2023-02-22 09:51:26 +08:00
liuhaozhe6788	6fac5490e2	将文本中的缩写转为读音	2023-02-21 10:48:12 +08:00
lhzstar	10f91f6c2a	修改输出音频路径并移动praat脚本到speed_changer路径下，提高了输出音频的音量	2023-01-04 15:25:11 +08:00
liuhaozhe6788	0612573975	将fixspeed.py脚本放入speed_changer路径中	2023-01-04 13:52:28 +08:00
liuhaozhe6788	b3be8d2a3b	命令行传参声音美颜权重	2023-01-01 21:02:14 +08:00
liuhaozhe6788	a96357d85e	demo_cli新增声音美颜权重参数	2023-01-01 19:48:20 +08:00
liuhaozhe6788	cd933eb354	修改合成器输出图表的路径名称	2023-01-01 16:20:34 +08:00
liuhaozhe6788	9ca22a72eb	修改合成器输出图表的路径名称	2023-01-01 16:16:24 +08:00
liuhaozhe6788	140b56a2e1	绘制所有文本片段的注意力对齐图和合成的梅尔频谱	2023-01-01 16:12:48 +08:00
liuhaozhe6788	498dbb6feb	给demo_cli.py中增加了更多文本预处理和清洗操作	2023-01-01 15:24:03 +08:00
liuhaozhe6788	0fb5bbe21f	不将文本的冒号改为句点	2022-12-21 18:43:30 +08:00
liuhaozhe6788	40175fb8ca	当标准音频不存在时，不执行声音美颜	2022-12-21 17:23:32 +08:00
WZY	1ecc4f5253	针对单句超长句的基于介词的文本分割,可以防止单句超长句文本出现崩塌现象.	2022-12-08 16:21:52 +08:00
liuhaozhe6788	bf2ddeb3dd	输入多段音频时显示当前输入音频个数	2022-12-05 13:31:12 +08:00
TiptoeHigher	54ebb84b48	merge the singal senteces to input	2022-12-01 10:27:37 +08:00
TiptoeHigher	a6db89370a	将分句录入语音合并后输入	2022-12-01 09:53:38 +08:00
liuhaozhe6788	f034998478	在demo_cli的主循环中，删除异常处理，显示运行错误的位置	2022-11-11 08:39:14 +08:00
liuhaozhe6788	bda5a670c1	文本划分中处理小数的特例，以及变速中处理输出音频速度为0的异常	2022-11-10 16:45:57 +08:00
liuhaozhe6788	699fb9f11f	add amp in demo_cli.py	2022-11-10 15:53:47 +08:00
TiptoeHigher	c0cf153c72	modified demo_cli	2022-11-04 17:26:20 +08:00
TiptoeHigher	5427b8239e	modified path in demo_cli.py	2022-11-04 16:49:26 +08:00
TiptoeHigher	9edd7c9ef1	initial commit	2022-11-04 16:38:13 +08:00