Multilanguage Support

[Intro: Multilanguage Support Made Simple]

What if your AI assistant could speak any language your customer uses automatically? That would be awesome. No extra setup, no hassle, just most conversation in their native tongue. Stick around because today we are making multilanguage support effortless. Wow, I can't believe it. In this lesson, I'll show you how to handle multilingual customers with ease. Whether you're using AI first automation or adding custom chat flows for specific languages, it's simpler than you think, but there is a few tricks to get it right. Let's dive in. If you want to use the simplest way to handle multiple

[AI-First: Let Your Assistant Detect Language]

languages, here's your golden rule. Go AI first and only. Why? Because your AI assistant automatically detects and respond in the customer's language. No extra settings required. For example, if a customer writes in French, the assistants will reply in French. If someone switch it to Spanish, boom, Spanish it is. This works flawlessly for most scenarios. But if you want to mix in custom messages or questions like free escalation forms or specific queries, things get a bit more hands on. And that's where chat flows come in. Let's say you want to add a custom process before ending off a conversation, like gathering extra details, or maybe you need different flows for escalation.

Here's the key. You will need to build a chat flow for every language you want to support. But wait, how do you make sure the right users enter the right flow? Well, you use filters. For example, if you want to target Italian speaking users, add this filter to the entry point, which is basically browser language is equal Italian. So let's see how an AI first and only approach will talk in Italian or talk in any language in the world without you doing any of the heavy lifting. Okay. So the setup is the same as the one we saw it and we're just going to speak Italian.

So basically for the ones that don't know what I did ask

[Custom Chat Flows for Localization]

I did ask I don't know how to issue an invoice. Can you help me? And now Aura is already recognizing my tongue. And that's it. That's Italian out of the blue. And basically the answer is correct. It's a perfect Italian as you can tell. Well, no, I can tell. I don't know if you can, but I can tell this is a perfect Italian. And here we go. Now, this will actually works also for intense. So if I go on now and try to match the intent that we defined before about pricing such as basically I have a question regarding the price for the invoices the additional invoices may know how much does it cost a thousand invoices per month.

Now if you see it she started writing something and then she stopped. Why? Because if you remember the last time we added an intention to escalate any pricing inquiries directly to the sales team and this is exactly what happened. Okay. So even if it was in Italian this was escalated to the sales team in the right way. Okay. And one thing that is cool about customer is that even though you don't understand Italian we make it seamless for you to understand the conversation no matter what. So this is fairly easy to quickly

[Fallback Flow Best Practices]

understand what they were talking about and intervene as well in your own language and then translate this in their own language. Okay. So this is the way aura will handle any incoming conversation no matter what either if it's a response coming from the knowledge base so the assistant mode or it's an intent with a mission. Now if we're going to ask again in another language, I'm sorry, I do know only two languages. But if I ask in any other language in the words, aura will just reply. So that's why you would use an AI first and only approach to cover any possible language in the world.

But if you do have to have specific flows to require different gathering of data which is not with AI missions as you've been seeing before then you will need to do a bit more of narrowing down for your travels. For example, let's say I want to narrow down this to only people speaking to a very specific language. I

[Why Multilanguage Builds Trust]

would add a filter here. Click property and then choose browser browser language which is it basically Italian. Now if I can save this now only the Italians will end up in this chat flow. Okay. So I can provide a very specific flow for the Italian speakers. And let's say if you got a team to assign this conversation whenever there is an escalation now you can escalate to the Italian support team rather than the general team. Okay. So that might be one of the customization but this is something that you can achieve in such different ways. Now you would need to defer the chat flows based on the browser language property only

[Your Multilingual Setup Checklist]

when you're adding messages custom messages that you want to use specifically like something like that. So this is something customized which require customization of course it requires a bit more of your time to set this up and this is why we strongly suggest you to go AI only because you can achieve the very same flow that you're doing over here with a mix of intense and AI mission. This is the old way to gather context to gather data from the user. Now you can gather context and data in a conversational way in their own language without creating a chat flow for every single language they've got.

But if you want to do it, it's completely up to you. The way to do it is like creating a chat floor for every single market or language. Now we're going to save this one. Okay. Now I need to tell you something. If you want to have several shuffles, one for each language, and then having a back support that goes to English, you would have something like this, but you need to exclude all the languages that you cover it with your dedicated chat flow. So at this stage if you go over several markets what we suggest you to do is to creating a list here adding over here a a property again language we're going to use browser language here what you could do is narrowing down all the browser languages that you cover specifically.

So let's say you want to have another one for Spanish and another one for French and so on. Now you need to change this instead of and it should be or because this filter needs to match whenever the pro language it's either Italian or Spanish or French. So now we can save this list and add other markets. Now that you've got this list, we should go back to chaflow. Now remember, you got the very specific one which will filter down conversations coming from people speaking that language. Now the general chat flow which will handle everything else should be excluding those languages. So you will add a filter on list.

You will choose the list that we created other market. And now you have to reverse this because you have to get over here all the people that do not belong to this list of other markets. So everyone else. Okay, this is the way to do it because when a conversation starts, it cannot be directed to two chat flows at the same time. A chat flow will be managing the conversation in real time and either ending off to another chat flow or ending off to your team or your teammate. This is why you need to narrow down for very specific languages if you want to go with customized messages or something very specific for each market.

Otherwise, if you want to go with AI first and only, then you won't need to do any of this stuff. Everything else will work just flawlessly. No need to add any selector for the language. The AI will recognize automatically for you the language and shoot the messages the right way no matter if it's an intent, no matter if it's a mission. So you can still gather data achieving the very same goal without freaking out basically and creating several chat flows, one for each market. But again, it's totally up to you. This is a suggestion. This is the way you can enter multi- language support. The inboxes customly helps you massively to manage multil- language support.

uh which you have over here the spoken language. So whenever the proen language isn't enough, you can also add the spoken language. When you're sure that your customers are speaking also another language which might be their main language as in my case I've got the browser in English but I speak mainly Italian. So I can put over here this is actually aura that set the Italian language when I started the latest interaction over here. So over here she understood that I was speaking Italian. She set the spoken language in Italian and now everything will be translated in Italian as well. Okay, this is the way you can manage multil- language conversations like in no time you can build flawlessly multil- language conversation thanks to aura genius.

If you want to manage specific flows then you will need to create several chat flows for every market or every language. Thank you so much for watching this video and see you in the next lesson.