From 989b755e8e9cfba8e8bb5fad7f672275980e1c1e Mon Sep 17 00:00:00 2001 From: WongLaw <95171490+WongLaw@users.noreply.github.com> Date: Fri, 9 Sep 2022 16:55:58 +0800 Subject: [PATCH 1/2] Revised must_neural_tone_words, test=doc. (#2370) * Revised must_neural_tone_words. --- paddlespeech/t2s/exps/vits/__init__.py | 2 +- paddlespeech/t2s/frontend/tone_sandhi.py | 5 ++--- 2 files changed, 3 insertions(+), 4 deletions(-) diff --git a/paddlespeech/t2s/exps/vits/__init__.py b/paddlespeech/t2s/exps/vits/__init__.py index abf198b97..97043fd7b 100644 --- a/paddlespeech/t2s/exps/vits/__init__.py +++ b/paddlespeech/t2s/exps/vits/__init__.py @@ -1,4 +1,4 @@ -# Copyright (c) 2020 PaddlePaddle Authors. All Rights Reserved. +# Copyright (c) 2022 PaddlePaddle Authors. All Rights Reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. diff --git a/paddlespeech/t2s/frontend/tone_sandhi.py b/paddlespeech/t2s/frontend/tone_sandhi.py index 9fff4272c..10a9540c3 100644 --- a/paddlespeech/t2s/frontend/tone_sandhi.py +++ b/paddlespeech/t2s/frontend/tone_sandhi.py @@ -30,7 +30,7 @@ class ToneSandhi(): '蛤蟆', '蘑菇', '薄荷', '葫芦', '葡萄', '萝卜', '荸荠', '苗条', '苗头', '苍蝇', '芝麻', '舒服', '舒坦', '舌头', '自在', '膏药', '脾气', '脑袋', '脊梁', '能耐', '胳膊', '胭脂', '胡萝', '胡琴', '胡同', '聪明', '耽误', '耽搁', '耷拉', '耳朵', '老爷', '老实', '老婆', - '老头', '老太', '翻腾', '罗嗦', '罐头', '编辑', '结实', '红火', '累赘', '糨糊', '糊涂', + '戏弄', '将军', '翻腾', '罗嗦', '罐头', '编辑', '结实', '红火', '累赘', '糨糊', '糊涂', '精神', '粮食', '簸箕', '篱笆', '算计', '算盘', '答应', '笤帚', '笑语', '笑话', '窟窿', '窝囊', '窗户', '稳当', '稀罕', '称呼', '秧歌', '秀气', '秀才', '福气', '祖宗', '砚台', '码头', '石榴', '石头', '石匠', '知识', '眼睛', '眯缝', '眨巴', '眉毛', '相声', '盘算', @@ -59,8 +59,7 @@ class ToneSandhi(): '下巴', '上头', '上司', '丈夫', '丈人', '一辈', '那个', '菩萨', '父亲', '母亲', '咕噜', '邋遢', '费用', '冤家', '甜头', '介绍', '荒唐', '大人', '泥鳅', '幸福', '熟悉', '计划', '扑腾', '蜡烛', '姥爷', '照顾', '喉咙', '吉他', '弄堂', '蚂蚱', '凤凰', '拖沓', '寒碜', - '糟蹋', '倒腾', '报复', '逻辑', '盘缠', '喽啰', '牢骚', '咖喱', '扫把', '惦记', '戏弄', - '将军' + '糟蹋', '倒腾', '报复', '逻辑', '盘缠', '喽啰', '牢骚', '咖喱', '扫把', '惦记' } self.must_not_neural_tone_words = { '男子', '女子', '分子', '原子', '量子', '莲子', '石子', '瓜子', '电子', '人人', '虎虎', From fdcc8c042762da5141fc5c59f43b37dfa28cfab7 Mon Sep 17 00:00:00 2001 From: TianYuan Date: Fri, 9 Sep 2022 17:49:20 +0800 Subject: [PATCH 2/2] Update README.md --- examples/aishell3_vctk/ernie_sat/README.md | 2 ++ 1 file changed, 2 insertions(+) diff --git a/examples/aishell3_vctk/ernie_sat/README.md b/examples/aishell3_vctk/ernie_sat/README.md index 777bea326..a849488d5 100644 --- a/examples/aishell3_vctk/ernie_sat/README.md +++ b/examples/aishell3_vctk/ernie_sat/README.md @@ -29,9 +29,11 @@ Or train your MFA model reference to [mfa example](https://github.com/PaddlePadd Assume the paths to the datasets are: - `~/datasets/data_aishell3` - `~/datasets/VCTK-Corpus-0.92` + Assume the path to the MFA results of the datasets are: - `./aishell3_alignment_tone` - `./vctk_alignment` + Run the command below to 1. **source path**. 2. preprocess the dataset.