Many companies require their voice applications to be available in more than one language. SAP NetWeaver Voice supports this capability with the Voice Kit.
If you want to internationalize your voice applications, consider the following points:
Provide the same options in each language required by your customers
Store the recorded audio files of different languages in a single directory
If the Text To Speech component of a third-party VoiceXML gateway supports multiple languages, you can use it as a fallback option to recorded prompts. This reduces the implementation cost of the voice application
A single voice application can support multiple languages. To enable this, you first need to define the languages that you want your voice application to support at application level and then define the multiple languages at the entity level.
Note
If your VoiceXML gateway is not multilingual and does not support Text To Speech then ensure that prompts for the various languages are available in the form of recorded audio files.
Note also that if the third-party VoiceXML gateway is not multilingual, the speech recognizer only recognizes input for the language it supports. It cannot recognize caller input in other languages. In such cases, you can use the DTMF option.
A voice application can use voice entities that are defined in a different language.
For example, you have a voice application with a component and have defined the voice application to use a male French voice and the component to use a female French voice. If control is transferred to the component and the defined audio files are unavailable there, the runtime environment uses the language defined at parent level.