IBM Cloud's Text-to-Speech Demo and Features (Full Transcript)

Speaker 1: Hello everyone, my name is Daniel Alvaro. This time we are going to show you a demo of what is text-to-speech, which is a resource that belongs to the area of Artificial Intelligence or Automatic Learning that IBM Cloud offers us. What does this service do? This service is practically to convert all the written text into a natural speech. The service transmits the synthesized audio with a minimum delay. In addition, the cadence and proper intonation are used for this process, corresponding to what is spoken and the language provided. It should be noted that within this text-to-speech service there are different languages ​​in which they are used. As can be seen later, there are different languages ​​such as English, Spanish, Italian, German, among others. In addition, it does not belong to a naive voice, that is, it is not a robotized voice. Here, despite the fact that it is an algorithm that does all this process, it is already dependent on what type of voice it chooses. If it is a male or female voice, it will have its attenuation and its corresponding difference to the voice of each one. In such a way that all this process of transferring all that is text-to-speech to a voice, to an audio, is much more natural. Next, we will show you the demo of how text-to-speech works by IBM. Text-to-speech As we can see, here we have the IBM Cloud platform. Here in the text-to-speech service we have a mega-summary. For example, here it says what type is this text-to-speech, this type of service, the provider, IBM, in which category it is located in artificial intelligence or automatic learning. We also mentioned that the plan that we chose, in this case it would be the free one, which would be the Lite, which is enough to carry out the activity. Within the text-to-speech as a resource as such, here it shows us the credentials and the URL that will be necessary to do the whole process. Within the initiation, it gives us a summary of what text-to-speech does, what it does, as it is known, it converts the written text that through a voice, being as natural as possible, says all this written text. In short, it is from the written text to an audio. Well, here it gives us some recommendations and what we are going to use, which would be the CURR, through a CMD or a Command Prompt. In this case we are going to use Windows, therefore we will apply the CMD. And here the syntax in which we have to refer to be able to carry out the practice. It is also mentioned that within the text-to-speech there are types of languages and voices with which we can use. For example, here we have the list of types of languages and genres and how they work. For example, we have Arabic, we have English from Australia, the United Kingdom, the United States, we have France, German, Italian, Portuguese, Spanish, Latin America and others. Once that is mentioned, we are going to do what is the practice in the three cases. These three cases are going to be English, which would be practically the Port de Pod that you have, with its destination voice and the text in English, as it should be. We have Spanish, in this case we chose, or it was practically chosen from Spanish from Spain, and the same text in Spanish. And the sentences, we made the change, instead of Spanish from Spain, we put the voice of Latin America. Now how does this work? It is to transfer all this code, it could be said, to the CMD. It should be noted that there are some things to consider, for example, in this part that says, header.asep.audio.slash.mp3 is going to be the format in which the file will be saved, and the output must match the format, in this case we put test1.mp3. Then what we do is copy this, we go to the CMD. It should be mentioned that according to the directory that we have, the files will be saved. As our directory is in desktop, which would be the desktop, they will be saved in our desktop. Then we copy and paste the code, and we wait for the test to be done. We go to the desktop, we can see that here it says test1. We double click. And as you can see, it transferred all the text to audio. Now, for the case of Spanish, we do the same. We go to the CMD, and we wait for the process to be done. We go to the desktop, and we do the test again, but now with test2.

Speaker 2: Hello, my name is Carlos Alfaro, and in the next audio, I am going to show you several famous phrases of famous important people around the world.

Speaker 1: And finally we have the phrases, which in this case would be Spanish and Latin American. We copy, paste, and wait. We do test3. And as you can see, all the translations were done correctly. Thank you.