Introduction - What we do

iSpeech, Inc. provides two types of services called Text-to-Speech (TTS) and Automated Speech Recognition (ASR). Text-to-Speech lets users to send us text and we provide those users with computer-generated speech. ASR allows users to submit audio data from a microphone or a file and we return a transcription of what was said.

Introduction - Types of services

iSpeech provides TTS and ASR services on various mobile, computer, and web platforms. For example, we support: Developers can build programs that use iSpeech services by using the correct iSpeech SDKs. iSpeech is capable of developing applications for paying customers who would rather pay us to create their app.

Getting access with API Keys

iSpeech's services are reachable with the correct identification code called an Application Programming Interface (API) key. The key is a token to access our TTS and ASR servers. Currently, iSpeech has 3 types of API keys with over 100 different options.

The three types of API keys are:

• Mobile Development
• Mobile Production
• Web/General API Production

The iSpeech mobile development server is used as a "sandbox" environment that you can use to test your application before releasing it to the public.

Building an Application - Mobile Platforms

In order to build a mobile application that uses iSpeech, you should download the appropriate package by using the Getting Started menu. The download will contain an iSpeech SDK library, documentation, and source code for a sample program.

The sample applications must be opened in the appropriate integrated development environment (IDE). For BlackBerry and Android, you should use Eclipse. For iPhone, you should use XCode. The sample applications can be executed as-is without changing the API key and server type.

Once you get the sample application working, try changing your API key and the production flags. In the BlackBerry application, open the Main package and the iSpeechSample class and set the following two variables.

Building an Application - Desktop/Server Platforms

The desktop/server SDKs for Java and .Net are only available on request. These SDKs are very useful for integrated voice recognition (IVR) phone systems. These SDKs require a general API key and are not compatible with mobile API keys.

You can create a general/web key under My Account -> My API keys. The web/desktop type API key requires credits which you must purchase.

Building an Application - Web Platforms

Many developers are interested in integrating TTS and ASR into websites. There are two main requirements for using iSpeech web services. The first requirement is that you submit the text for TTS or audio data for ASR to the iSpeech servers using the HTTP protocol. The second requirement is that you must have a way to receive and present the data to the user.

For TTS by default, you will receive an MP3 that you can play with flash and/or javascript. For ASR, you must use a browser plugin that has access to the microphone such as Flash or a Java applet.

Optimization

For the web platform, you can use HTTP GET requests rather than POST because it requires fewer packets. However, GET is limited by maximum URL length.

You can improve download speeds and save credits by downloading one sentence at a time in a playlist. The time that it takes to convert one sentences causes natural sounding short pause.

Troubleshooting

The frequently asked questions (FAQ) page is available to answer common questions. For other questions, you can ask your question by posting on the iSpeech Forums or with the support contact form.

Your application will require an internet connection. If there is a problem with your internet connection, iSpeech services will become unreachable.