Apps|Java Speech API

Hi Ness,

The Quickstart guide should have the info you are looking for, it can be fould here...

// This quickstart guide requires the Java codekit, which can be found at:
// https://github.com/attdevsupport/codekit-java

Good luck!

-Bill

Hi,

Thanks for your help.

Finally I completed the first step, I can send wav file and get text, where is the documentation to use chunked, on live speech to text?, and also for text to speech?,i was looking for examples but i not found nothing.

In my first question i was using netbeans was not possible to create the .jar file, i did it manually, installing maven and the set as java 6 as principal JDK and then run manually the command for maven to create the .jar file.

Hi, Again,

Now i can convert text to speech the example is here:

https://github.com/attdevsupport/codekit-java/blob/master/examples/tts/src/main/java/com/att/example/App.java

Also i can convert text from any language to speech just changing this line of code:

byte[] audio = ttsService.sendRequest("text/plain", "Mi nombre es Rosa y hablo espanol al fin.", "VoiceName=rosa,Volume=100");

VoiceName= is the important thing to request the speech idiom.

English: crystal,mike,etc.

Spanish: rosa, alberto,etc.

Only two more thing is missing:

1. How can I convert Speech to text on live? Spanish and English
2. Where is all the documentation or tutorial or more example with explanation or a book?

Thanks to all for all your help

Hi Ness,

Here is a snippet taken from the "AT&T¿ Speech SDK Developer Guide for Android" doc

/*The Speech API recognition service supports clients that use HTTP streaming when posting audio data to the service. In a streaming request, the service performs speech processing while receiving data from the client. By contrast, in a non-streaming request, the service waits until it receives all data before performing speech recognition.

A client signals a streaming request to the server through a standard HTTP 1.1 mechanism called chunked transfer encoding. Adding the Transfer-Encoding: chunked header marks a streaming request. When the streaming request is signaled by adding this header, the client segments the posted data into chunks; a chunk size of 512 bytes is good for speech recognition.

For more details on chunked transfer encoding, see the following HTTP specification at the following location: http://tools.ietf.org/html/rfc2616#section-3.6.1

If the client does not include a Transfer-Encoding header, it must include a Content-Length header instead, which specifies the number of bytes being sent in the request.*/

This doc is included here: https://developer.att.com/static-assets/packages/SDK-IDE-Libraries/ATT-API-Platform-AndroidSDK-1.2.1.zip

Does this answer your question?

Thanks

-Bill

go to API downloads section

this one is good i have tried something like this on a client's app.

https://appdevelopmentagency.com.au