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.
French Canadian Harmonie, need some more tuning. Does not sound that great compare to acapela or RealSpeak
Hi, Denis :
Thanks for the feedback. Do you have some specific comments from intelligiblity, prosody naturalness or sound quality perspective?
How do I create SR and TTS for my language? Beside our national language, we have many region languages. It would be nice if we can preserve the languages. thanks for guide.
For your region language, do they have different lexicon or pronunication system?
Seems that you want to create dialect voice, there is not public tools to do so yet. But I think this is a direction of TTS in the future.
Wait this is for Server OS only? I installed it on Windows 7 but no voices show up in the Text to Speech control panel.
Why not make these server only voices available on the desktop?
These voices are optimized for server scenerios.
For example, the output format of the voice is 8k16bit.
But you can use the voices on desktop by writing an app with sdk but not with the text-to-speech control panel.
yes, it is difference lexicon and pronunciation system per region languages.
Since it own region has they languages and culture, just like mini world.
Hope we can get the sdk. thanks for explaining. keep a hope. :)
Can you explain what this package is? How does it relate to UCMA? Is it a new package, and if so why is it version 10.1?
I think this is great news for us. However, I must admit, I'm confused. Can you please do a post that explains the difference between this new "Speech Platform", the UCMA 2.0 SDK, and the Microsoft.speech and System.speech namespaces?
I think this new release takes the Speech features from UCMA and the Recognizer and TTS engines from OCS and makes them available for reuse. So, with this I can build server speech apps without OCS. Is this correct?
UCMA SDK is with OCS. Now speech platform is a standalone SDK
Does this require us to buy a separate license? I have a TTS synthesis application running on a server and would like to replace the Desktop Speech to Speech Server.
From technology point of view, you can certainly do this.
For license issue, I am not a lawyer, but you can check the license agreement to see if it matches your scenerio and need.
Can't get it going.
1. Running Windows 7 Pro x64 with Visual Studio 2008.
2. Installed Runtime x86, SDK x86 and SR en-GB.
3. In VS set platform target to x86 for All Configuration
Copied and pasted the code from documentation file:
private void Form1_Load(object sender, EventArgs e)
// Create a new SpeechRecognitionEngine instance.
sre = new SpeechRecognitionEngine();
// Create a simple grammar that recognizes “red”, “green”, or “blue”.
Choices colors = new Choices();
GrammarBuilder gb = new GrammarBuilder();
// Create the actual Grammar instance, and then load it into the speech recognizer.
Grammar g = new Grammar(gb);
// Register a handler for the SpeechRecognized event.
sre.SpeechRecognized += new EventHandler<SpeechRecognizedEventArgs>(sre_SpeechRecognized);
// Simple handler for the SpeechRecognized event.
void sre_SpeechRecognized(object sender, SpeechRecognizedEventArgs e)
Program compiles but doesn't recognize my voice.
Hi, thanks for your post!
Well, i have been trying to use this in C#, i'm importing the COM Microsoft Speech Object Library version 10.0 (guess that's the one that belongs to this SDK). I'm programming under Windows 7 OS, with VS 2010.
Ok, so, i'm just trying to use the next code:
SpVoice voz = new SpVoice();
But, it throws an exception: HRESULT: 0x8004503A
Don't know what i'm doing wrong, could you please help me?