Fix Naver Papago

• fixed naver papago

Voice list update, fix for SSML characters

• update voice list, there are some new Azure voices
• enhance fix for SSML characters on Azure/Amazon (<, >, &)

CereProc service, fix for HTML special characters

• CereProc TTS has been added, which brings a number of English voices as well as some European voices.
• There is now an option to escape HTML special characters such as <, >, &, which maybe be necessary in certain technical decks such as medical, where previously sentences like "patients age < 30" wouldn't generate any audio on SSML services such as Azure.

Azure: support mstts SSML tags

Azure can now handle mstts SSML tags, such as <mstts:silence type="Sentenceboundary" value="200ms"/>
full list of possible tags here: https://docs.microsoft.com/en-us/azure/cognitive-services/speech-service/speech-synthesis-markup?tabs=csharp

Voice list update, fix FPT.AI Linh San (Ace) voice

• Voice list update, fix FPT.AI Linh San (Ace) voice
• added switzerland region for Azure

Fix FPT.AI bug when using user keys

• Fix FPT.AI bug when using user keys

Updated FPT.AI Vietnamese voices

• Updated FPT.AI Vietnamese voices

Updated Naver voice list, fixed Forvo / Mandarin issue

• Updated Naver voice list (new Korean and Japanese voices)
• Fixed a bug when using Forvo with Mandarin and AwesomeTTS Plus

Updated voice list (Arabic and Spanish)

• added a large number of Arabic and Spanish voices from Azure

Audio cache directory, Naver Papago fix, Forvo fix

• Audio cache directory moved to user_files to avoid being overwritten after an upgrade
• Naver Papago fixed.
• Forvo now looks at a few top search results to more accurately find desired words (contributed by TiddlyWiddly)