Thank you very much, Stephen and folks. It's good to be here tonight to talk about this new feature. We're really excited about it. And the feature is called Global Sermons. Global Sermons. And what's the idea behind that? Well, many of you who are broadcasters here tonight faithfully preach and upload your sermons every week into Sermon Audio, or you have someone to help you do that. And we receive those sermons in audio file format or video format. And we take those and process those, and if you preached your sermon in English, well then, once those are available on our website, then people from your local assembly or even all over the world who speak English are able to benefit and be blessed by those sermons. So wouldn't it be neat though if there were a way where we could take your English language sermon and then make it available somehow to people in other language groups, Spanish or Italian or any one of a number of different languages where they could go to the sermon audio site Go to your sermon and hear it preached to you, to them, in their own tongue. And so that's the idea behind this feature, global sermons. And so we're launching it earlier today, we did, and so if you're a broadcaster, well then you can go to your broadcaster dashboard, you're familiar with that, or maybe there's someone in your church who helps you with that. They can go there and they can see this new feature and take advantage of it. And we hope it'll be a blessing to you and to the Lord's people. So what is it all about? Well, it works this way. First of all, it starts with your upload, your upload of your sermon in audio file format. And then we take that sermon and then it goes through kind of a multi-step process. And each step in that process uses kind of an emerging technology that I'll talk about in just a moment. I call it emerging, but actually it's been around for quite a while. But it's only in recent years that this technology has really accelerated in its development. It's become widely known. In fact, if you've been following the computer world, you can barely turn around without bumping into a new feature powered by this technology. And what I'm referring to is something called artificial intelligence or AI. Maybe you've heard of that. There's probably many people here that are familiar with that. And this technology, a lot of people are asking different questions about it. Is it safe to use? Is it ethical to use in a religious organization or an educational institution? Is it going to take my job away? People have concerns about that technology. And so the question faced to Sermon Audio is, well, can we use this technology in a way that honors the Lord and that is a benefit to his people? And well, we think we can do that. And we're implementing some safeguards to kind of ensure the integrity of the process. So this technology, AI, it turns out it's really good at dealing with languages, and that's kind of what we're dealing with here with sermons. So let me describe the three different steps of, to coin a term, globalizing a sermon. That is taking a sermon in an original language that was preached in, like English, and then making it available to be listened to in other target languages, like Spanish, French, or German. or any number of different languages. We have 46 different languages that we can globalize these sermons into. So the first step in the process is you uploading that sermon. That happens every week. You do that already. And the first step is called a transcription. Transcription. And what happens there is we take your sermon and we feed it into these artificial intelligence engines. We call them models. And what it does is it dissects your sermon, listens to the audio, so to speak, and it divides it into its constituent parts of words, sentences, and paragraphs, and then forms kind of a textual version of your sermon, kind of a transcript. And that feature actually is already available today in Sermon Audio. If you go to any sermon or most sermons, you can see a transcript, a printed transcript of that sermon. You can print that out, you can read it, you can give it to people. And it's kind of handy as a record or a reference of what you preached. But that's the first step in globalizing. And then we require of you, you or someone who works with you, to actually look and review that transcript in the sermon manager. We want to make sure that before we move on to the next step, that what's recorded there in that textual transcript is actually what you intended to say. So you'll have a way of reviewing that. And then you'll have a way eventually of approving that before it goes on to the next step. So that's kind of one of the safeguards of the human oversights we're employing to make sure that there's integrity to this process. And it's not saying something we didn't want to say. So that's the transcription portion of it. And once you approve that, it then goes on to the next step. Before you do that, you can actually look at the transcript. You can modify it. We have a little sub-feature that we've introduced, something called Smart Edit, where you can click on the Smart Edit button for your transcript, and it will feed that transcript into the artificial intelligence engine, the model. And it'll ask it to review the transcript and maybe offer some suggestions. Are there any run-on sentences or any fragments or anything that don't follow the rules of grammar? And it'll identify those things to you, and you'll be able to look at those. And it'll offer suggestions, and you can accept those or reject those. And then once you're happy with that kind of finalized transcript in a more pristine state, then you can give your approval to that, and it'll move on to the next step of the process. So that's the transcription. part of the process. Step number two is the translation. So we move from transcription to translation. And the translation part of that is, again, using AI. And we're taking that transcript that you've approved, and we're feeding it into the model. And we're asking it, at this point, to translate it into the target languages that you have specified in your servant manager. You may have preached it in English, but you may say, well, I'd like to have that available in Spanish, Italian, and German. And so at this step, it'll take that English language transcript, and it'll review it, and it'll use all of the information that it has at its disposal to actually translate this into that target language. Now it's interesting, these models, they call them large language models, and they call them that for a reason. First of all, they're large, they require a lot of computing capacity and power to execute, but it turns out they're pretty good at language as well. They've been trained with information from all over the world, all over the internet. You'll have documents, articles, poems, manuals, pieces of literature. All those things have been provided to these models to train them, okay? And they systematize them and they organize them in a way such that later on when you ask these models a question or when you ask them to perform a task like translation, it will have all that information kind of at its fingertips to perform that task. It'll have examples of the different languages and examples of translation. So it'll be well-equipped to do that. And so this is not kind of a word-for-word substitution. This is really kind of an intelligent, if I could use that word, translation. And when you look at the result, you'll actually be quite amazed. It looks like something that somebody actually wrote from scratch. And we've had people review this process with us, and some of the feedback that we've been getting are amazing in words like uncanny, Well, you can be the judge of that when you try it yourself. And if you're a broadcaster here tonight, you actually have something in your dashboard, a coupon, that you'll be able to go and use this feature. Try it out even tonight. So anyways, we've moved from transcription to translation. And we have this initial translation that was done with the transcript. And what's the next phase? Well, maybe you experienced this when you were a child. I did. When I was in my class, either my math class or my writing class or English, my teacher would inevitably say to me, Go back and check your work. So that's what I would do. I would go back and check the paragraph. Inevitably, I would find something wrong there. Maybe there was a run-on sentence, or maybe I violated some rule of grammar, or I made some mistake in my math problem. And I would check my work, and I would find and correct those things that I needed to correct. That's what we're doing here as part of the translation process as well. Once we have that initial translation, well, we'll take it along with the initial transcript, and we'll feed it back into the AI engine. And we'll ask it to do a review. We'll tell it to check your work. It will go back and review the translation that it produced. And it will find suggestions and corrections. And actually, we found that it's been pretty good at producing a translation that was better than the one that was produced initially. And you might ask yourself the question, well, why didn't it come up with the corrected one to begin with? Well, that's a good question. And there are some computer science people here that you can ask that question tonight. And they can probably answer that question. But what we found is the more that we do that, the better the translation becomes. The second step will take that transcript and the translation, feed it into AI, and then come up with a better translation. Finally, the third step of that translation process will be to take the final translation, put it one more time into the AI model, and then it will come back with a smoother translation. We will inform the model. We will give it a set of instructions, and they call those a prompt. We will tell the model, please Act as if you're like a French speaker, and this is what you're about to say, and please smooth it out and make it correct in terms of speaking. And it does that, and the end result of this process is we have a translation that's pretty good. And at that point, here's again where we engage some human oversight. So at this point, if you go back into your sermon manager, that translation will be available for you to review side by side with the original transcript. Maybe you have someone who works with you or works in your church, and maybe they speak Spanish, for example, and they'll be able to review that Spanish language translation. And they'll be able to compare it to the English transcription and make any alterations or corrections necessary. So once they give their approval to that process, to that translation, then it goes on to the third and the final step, which is what we call voicing. So we've got transcription, translation, and voicing. And here's the part where it gets exciting. We take this translation into this target language and we provide it again to the AI model and we'll ask it to read it aloud in the language of the target language, Spanish or Italian or whatever it may be. And we'll capture the output of that and we'll put that back into sermon audio so that when someone goes to the website and they visit your sermon, they will of course see the original English language version, but they will also be able to choose from a series of languages that you've chosen and they'll be able to click on that language and have the sermon then read aloud to them in their own language. And so that's the goal of these three steps. So we hope it's a blessing to the Lourdes people. We know that by providing your sermons to people in different geographic locations, you're really extending the reach of your pulpit. And by providing this feature, we hope that it will be extended even further. by different people, groups, and different languages that ordinarily wouldn't be able to listen to your sermon. So that's the idea behind it. And at this point, we have a little video that we'd like to show you that will kind of demonstrate how it works. And we hope it's a blessing to you and to the Lord's people. So John, if we could have the video straight away. Hello, my name is Stephen Lee. I'm the founder of Sermon Audio. And we are launching global sermons with 46 languages. This is a very exciting day for us because we've been working on this back in 2023. And the idea was floated of taking an English sermon and having it re-preached in multiple languages. We even took this concept of an idea to Africa, where we wanted to take this globalized sermon from English to Swahili and see what the local folks thought about that. And it seemed to work. And here we are today. We couldn't be more excited. Here's how it works. Let's take a brief look at the Sermon page for a Global Sermon. Next to the title, you'll find an added language selector. When a user chooses their preferred language, the player's content will be replaced with a translated audio version in that language. There are a number of visual cues to indicate to the user that they're looking at a translated sermon rather than the original sermon. The page is given a blue border, the audio player turns blue, and translated text is marked with a blue globe icon. Any sermon on the site can be converted into a global sermon. To convert one of your sermons, navigate to the Sermon Manager in the Broadcaster Dashboard and open your sermon in the Sermon Editor. Everything pertaining to sermon translation can be found in the Globalize tab. When you upload your audio or video file, our system generates a transcript of your sermon's content. we require that broadcasters review and approve the source transcript. Once approved, the source transcript text is then automatically translated into whichever language you've selected. From this interface, you can select multiple languages to translate to, as well as the voice to use for each audio translation. At this stage, we once again give broadcasters the opportunity to review the translated transcript before proceeding to the final step. Click this Review button on the transcript to open the transcript editor where you can make revisions to the translated transcript. Once you've reviewed the translated transcript, click the Approve button. Once the translated transcript is approved, the system will generate the spoken audio version for the selected language. Now that the voiced sermon has been generated, the new language audio will be available for users to listen to on the sermon page. We believe that this implementation of taking sermons from one language and having them available in multiple other languages is the strongest, most reliable implementation that there exists because you have three safeguards that we've built into the whole process. Number one is smart edit. Smart edit is a kind of a cleanup step, removing any faux pas that may be common in extemporaneous speaking. like repetitions or subject-verb disagreements or sentence fragments, and even flattens out idioms so that it prepares the text for the best possible translation. The second thing is the AI process itself. We're working with the best models that are on the market today, but we are doing so in a way where there are multiple steps within the actual translation process that will mitigate some of the issues that we find with artificial intelligence. And the third safeguard that we've integrated, human oversight. We make it very easy for there to be human reviewers. You can have multiple users that are assigned in a translator role that can log into the system and see only their work to be done. So with these three safeguards, we feel very strongly that this is an incredible effort. We are giving churches the ability to reach the world. And it's not just from the English-speaking world to the rest of the world. It's from any language reaching Any language, whether it's in a small church in America, or a house church in France, or a small congregation in Korea, or even a preacher in Africa. Any church in any place can reach anyone in any language. It's incredibly exciting. Acts 2 verse 11 comes to mind. We do hear them speak in our tongues the wonderful works of God. Exciting, right? It's really exciting.