While researching how to add text-to-speech content to my websites, I've learnt some important things. It's about the root cause of what's holding us back from real innovation.
Yesterday, my friend gave me some valuable feedback on my content. He's a nice person with years of experience in UX design, and we used to work together in Southeast Asia. If I need to build a web or a mobile app from scratch, or start a small business in the region, I want to work with him again.
His advice is to add audio content to my websites. Text-to-speech and podcasts are very popular. The majority of people don't read text content and prefer to consume video or audio content. My websites are optimised for tablets like the Kindle HD for reading text content, but it's also a good idea to add audio content.
Last night I did some research on how to add text-to-speech in a cost effective way. Because my websites are run entirely as a non-profit and open source project, so I don't have a budget for this. All I can fully invest is my consciousness, experience, time, knowledge, techniques and a small amount of money.
Your support is always welcome through Buy Me a Coffee! ☕️
And the most important thing for me is to care about privacy and security for sovereignty and freedom. This is my core mission to work on this project, so my research was also focused on low cost, easy to use, privacy and security for text-to-speech stuff.
From February to May I was involved in a joint venture project. The project aimed to automate time-consuming tasks and complex workflows that frustrate CEOs and small teams. During the process of designing the product and solution, I built a lot of small prototypes to make sure what worked and what didn't.
In some prototypes, automating workflows in marketing and business operations, I integrated some AI to work with text, audio and images. Specifically, I used the OpenAI API to generate text, the Google API to generate speech, and the Stable Diffusion API to generate images. The prototypes then sent a customised message to the audience.
It's not hard for me to build a custom workflow for creating audio content with NoCode tools, but I've found it extremely hard to create audio content with AI while considering cost and privacy.
Even though some great developers share their valuable work as open source in the public repositories, it requires a powerful machine and Linux environment. I used to use Debian a lot, from Woody to Jessie, so I tried to install and use a virtual machine on my laptop to use the Linux environment.
But I had to go through so many steps to get things done. It really felt as if the deception systems were made up of multiple illusions and countless mind cages as nested virtualisation. Logical and physical distance creates walls, including paywalls, that prevent me from making things happen.
That's the root cause that has kept us from real innovation, as we explored in the previous insight. We're in a giant paywall called the money system and we're in the illusion system of money for power and interest game of the global elite and the human reps of the ancient deities - the Anunnaki slaves.
I fully agree with that paid services focus on real privacy and security, while many freemium services steal and sell users' information. That's the main reason why some big tech companies have recently put paywalls on their freemium. They are trying to keep out the harmful bots that cause fake news and misinformation.
Yes, as you've already noticed, we're in a super suffering paradox because of the money system. To change this painful situation, we first have to make a firm decision to stop being enslaved by others. Stop serving the deceptive deities. Stop working for the global elite and the human reps in the back room.
When it comes to my websites, Softr doesn't allow me to embed content with a free plan. And I can't pay 59 USD a month for the website. So I have to be more creative about how I add more 'must have' stuff to the existing project and what are the 'nice to have' things I have to give up.
If the audio content generated by text-to-speech technology is necessary, I can create an audio generation workflow using Make.com and Google API. Because my content is public and open source. When I use your content and messages, I have to be very careful about your privacy.
As we've explored today, a little challenge always gives us a huge awareness.
P.S. I've found a very good solution for storing audio and images. Like the image above, I've already used it to store and serve the images in this 'short insights' section. It's Cloudflare R2 (object storage in the cloud), and it doesn't require any programming skills to store content. If you use R2 with a custom domain like I do, CDN (content delivery network) is automatically available.
P.P.S. If you're interested in automating your tasks and workflows, Make (formerly Integromat) is one of my favourite solutions for doing this very quickly and inexpensively. If you sign up through my affiliate link, you'll get a 1 month free option. This option gives you 10,000 operations for free, while the free plan gives you 1,000 operations per month. That's 10x the value for you!