Fine-tuning TTS With VoiceMarkupKit

Text-to-speech (TTS) markup is text with imbedded indicators that control speech synthesis from the text. Speaking qualities such as the speed, pitch, emphasis, and word pronunciation may be tailored in reproducing speech from text.

Chant VoiceMarkupKit is comprised of software class that handle the complexities of generating text-to-speech markup for various markup syntax. This enables you to tailor speech synthesis to produce sounds in familiar dialects, speaking patterns, and accents of your end users. You can adjust TTS markup as needed for the synthesizer to enhance the playback quality when synthesizing.

Synthesizers (i.e. speech APIs) interpret different markup syntax. VoiceMarkupKit supports the following markup syntax:

Speech API	Markup Syntax
Acapela TTS	AcaTTS Tags
Cepstral Swift	W3C SSML
CereProc CereVoice	W3C SSML, CereVoice Tagset
Microsoft Azure Speech	Azure Speech SSML
Microsoft SAPI 5	SAPI 5 XML Markup, W3C SSML (SAPI 5.3+)
Microsoft Speech Platform	W3C SSML
Microsoft .NET System.Speech	W3C SSML
Microsoft .NET Microsoft.Speech	W3C SSML
Microsoft WindowsMedia (UWP and WinRT)	W3C SSML

For more information about the fine-tuning speech synthesis with VoiceMarkupKit, review the following topics:

• Text-to-Speech Markup and
• Synthesizing with Markup.