text-to-speech

Specs

• Leon version: latest
• OS (or browser) version: Fedora 30
• Node.js version: 10.16.3
• Complete "npm run check" output:

➡ Here is the diagnosis about your current setup
✔ Run
✔ Run modules
✔ Reply you by texting
❗ Amazon Polly text-to-speech
❗ Google Cloud text-to-speech
❗ Watson text-to-speech
❗ Offline text-to-speech
❗ Google Cloud speech-to-text
❗ Watson spee

目前的多音字使用 pypinyin 或者 g2pM，精度有限，想做一个基于 BERT (或者 ERNIE) 多音字预测模型，简单来说就是假设某语言有 100 个多音字，每个多音字最多有 3 个发音，那么可以在 BERT 后面接 100 个 3 分类器（简单的 fc 层即可），在预测时，找到对应的分类器进行分类即可。
参考论文：
tencent_polyphone.pdf

数据可以用 https://github.com/kakaobrain/g2pM 提供的数据

进阶：多任务的 BERT
![image](https://user-images.githubusercontent.com/24568452

Hello,
I tried both espeak-ng-20191129-b702b03-x64.msi and espeak-ng-20191129-b702b03-x86.msi
as well as building the project from the source code but the Arabic language isn't showing up in all cases. The build was successful but the Arabic language doesn't show up when I try to check "espeak-ng --voices"

checkinstall scons install

gives an error:

Install file: "build/linux/lib/libRHVoice.so.5.1.0" as "/usr/lib/libRHVoice.so.5.1.0"
scons: *** [/usr/lib/libRHVoice.so.5.1.0] No such file or directory