Want to build your speech application with speech recognitions and synthesis?
Install the package below, you can now use SR and TTS in 26 language easily.
The 26 languages are:
Chinese (Hong Kong)
Feel free to try the TTS in the 26 languages, post your feedbacks about them here.
do you have sound card on your windows 7?
Don't know if the question is for me, but, if that's the case, yeap, i have sound card
Do you know if there are male voices available in the runtime? If not, are there any plans by the MS Speech team to add male voices in the future?
Currenlty no male voice.
We will consider the male voice in plan.
What exactly do the language packs indicate? If I am using the Chinese language pack, does it mean that all english text is translated into chinese and converted to speech or does it pretty much speak English text in a Chinese "accent"?
Language pack means it reads the text in that language like a native.
Chinese language pack means a Chinese TTS.
@myblog - I am not sure if I understand correctly. If I use English language text and use a Chinese language pack, does it read the English text in the Chinese accent? What's the correct way of using the non English language packs?
>> We will consider the male voice in plan.
Thanks. Do you have an estimate on when this support will be added to the runtime? (Don't need an exact date, just want to know if this might take months or years)
Our Chinese language pack can read English word inside it.
Usually you use a language pack if your incoming text is mainly in that language.
Sorry, I cannot give you the ETA of it for now. But your feedback is heard.
I'm starting to research about SR platforms and can't get it running for Portuguese (Brazil). Where can I find some samples? Tried all available in samples/winui/speech Windows SDK folder but no success.
Have you looked at the SDK package?
For the samples in Windows SDK, they are for desktop speech API.
Here is the difference between desktop Speech API and server Speech API:
1. They have different GUIDs for COM object.
2. They have different name space. (System.Speech for Desktop, Microsoft.Speech for Server)
Is there a way to process an AMR (Narrowband) audio as an input for the speech recognizer? It seems I can only use PCM. Since AMR is smaller I was hoping I could use it since the speech I am recognizing comes from a recording made in a phone.
i installed speech run-time in my system and Japanese language pack also but when i tried to convert Japanese text in voice then the engine is not speaking synthesizing to voice so can u please tell how exactly i need to follow. and is there any case like i must have Japanese voice then only engine will synthesize the text?
Can I use the new Microsoft Speech Platform with MsAgents ? It works fine ?