Google Cloud Text-to-Speech API supports SSML.
So, I would like to add an option to specify Plain Text or SSML to Say WaveNet.
https://cloud.google.com/text-to-speech/docs/reference/rest/v1/text/synthesize#SynthesisInput
I understand.
If "<speak>%text</speak>" in the Text field, it is interpreted as SSML.
However, if "%text" is set to "<speak>test</speak>", and the Text field is "%text", it is interpreted as Plain Text.
Should both be interpreted as SSML?
Thanks.
Both are interpreted as SSML. There is no difference for those 2 use cases :)
A1: Variable Set [ Name:%text1 To:test1 Recurse Variables:Off Do Maths:Off Append:Off ]
A2: Say WaveNet [ Text:<speak>%text1</speak> Voice:en-US-Wavenet-A Stream:3 Pitch:15 Speed:8 Continue Task Immediately:Off File: Override API Key:********** Respect Audio Focus:On Continue Task After Error:On ]
A3: Variable Set [ Name:%text2 To: <speak>test2</speak> Recurse Variables:Off Do Maths:Off Append:Off ]
A4: Say WaveNet [ Text:%text2 Voice:en-US-Wavenet-A Stream:3 Pitch:15 Speed:8 Continue Task Immediately:Off File: Override API Key:********** Respect Audio Focus:On Continue Task After Error:On ]
No, there is a difference.
test1 is interpreted as SSML. In other words, it says "test one".
test2 is interpreted as Plain Text. In other words, it says "less than speak greater than test two less than speak greater than".
Thanks.
Do you maybe have an extra space before <speak> in %text2?
Yes!
I'm sorry, I did not notice that there was a space.
The criteria to judge SSML are a little too strict, but I understand the criteria, so this is no problem.
Thank you very much.
You can already use SSML :)
