Oracle Bot comes with a very sophisticated language detection and translation service. You can attach either a Microsoft or Google service to the environment and make use of it within your bot. Oracle Bots dev guide talks about it : https://docs.oracle.com/en/cloud/paas/mobile-suite/develop/localization.html#GUID-997ED305-D011-4E89-9440-566092A4870A
But, these translation services are not free. So for my development purposes I created a custom component, which does the same for me but for FREE.
Introducing Yandex APIs (https://tech.yandex.com/translate/). It is simple and easy to use. And most importantly it is free.
Yandex supports over 90 different languages. The only thing you need to do is, register and create an API key for yourself.
Step 1:
Register and create an API key : https://translate.yandex.com/developers/keys
Step 2:
Download the custom component from my GitHub : https://github.com/sohamda/CustomComponents/tree/master/detect_language
Step 3:
Update the "detect_language.js" and add your key to the detectLanguage() function.
Step 4:
Update "registry.js" to add this new custom component.
Step 5:
Add a string variable in the dialog flow, named "detectedLocale".
Then update the dialog flow to use this custom component, followed by a "Switch" component in order to decide what to do once you detected the language.
You can use as many as transitions you want in your switch. In the above example I just did two.
Once this is done, test your bot.
That's it!.
Happy "Bot"-ing. :)
But, these translation services are not free. So for my development purposes I created a custom component, which does the same for me but for FREE.
Introducing Yandex APIs (https://tech.yandex.com/translate/). It is simple and easy to use. And most importantly it is free.
Yandex supports over 90 different languages. The only thing you need to do is, register and create an API key for yourself.
Step 1:
Register and create an API key : https://translate.yandex.com/developers/keys
Step 2:
Download the custom component from my GitHub : https://github.com/sohamda/CustomComponents/tree/master/detect_language
Step 3:
Update the "detect_language.js" and add your key to the detectLanguage() function.
Step 4:
Update "registry.js" to add this new custom component.
'DetectLanguage': require('./detect_language/detect_language')
Step 5:
Add a string variable in the dialog flow, named "detectedLocale".
detectedLocale: "string"
Then update the dialog flow to use this custom component, followed by a "Switch" component in order to decide what to do once you detected the language.
detectLanguage: component: "DetectLanguage" switch: component: "System.Switch"
properties: variable: detectedLocale source: values: - "en" - "nl" transitions: actions: en: <ENGLISH_STATE> nl: <DUTCH_STATE> NONE: <EXCPETION_STATE>
You can use as many as transitions you want in your switch. In the above example I just did two.
Once this is done, test your bot.
That's it!.
Happy "Bot"-ing. :)
Chatbot Development in India, USA, UK,
ReplyDeleteChatbot Vendors in Lucknow India,
Hire Best Chatbot Developers in Lucknow India,
Chatbot Development in Lucknow India,
Chatbot Builder Lucknow India,
It was an amazing blog.
ReplyDeletechat bot artificial intelligence