Valentino Guerra

🔉 Add a Voice to Your Articles with AI using ElevenLabs

June 11, 2023
Add a Voice to Your Articles with AI using ElevenLabs

It’s a reality, and you can do it right now!

The day has come when you can use artificial intelligence to give voice to your creations in a way that is indistinguishable from a human. Today I’ll show you how to achieve this.

It will only take 4 simple steps, and in less than 15 minutes, you’ll have excellent quality results.

Step 1 - Convert the article to plain text ✨

Transition between formatted text to plain text
Transition between formatted text to plain text

First, select and copy all the text from your article with Ctrl + C (⌘ + C on Mac). Once copied, we need to convert the content into plain text, meaning that it should not contain symbols (#, <, -) or any syntax specific to the text editor we are using (Notion uses Markdown). This way, the AI can read it more easily.

To resolve this issue, I've created a small, free tool you can use: TextToReadable.

Step 2 - Generate audios with ElevenLabs 🎙️

Right now, ElevenLabs is the best option available, both for the quality of its output and its low costs. It even offers a decent number of free generations that renew monthly.

You can choose between using ElevenLabs or any other reliable option, here is a comparison between ElevenLabs and

In this example, we are going to use ElevenLabs. First, navigate to Speech Synthesis. There, choose the voice that you like the most and that suits the tone of the article you are writing. You can also experiment with additional settings.

Screenshot of Speech Synthesis from ElevenLabs
Screenshot of Speech Synthesis from ElevenLabs

Once configured, simply paste the text you want to convert to audio. To convert the entire article, we'll generate the text in chunks. Each block should be less than 10,000 words, but I recommend limiting it to 2-3 paragraphs per audio we generate. This way, if we need to fix or change something later, we'll only have to regenerate that specific audio block, not the entire article.

Paste the first chunk of paragraphs into the box and press generate. Then, download the generated audio.

Rename the file from ElevenLabs_2023-07-21T16_51_06.000Z_Rachel_OuKa9BSoXUpHNwKducjI.mp3 to 1.mp3so that is easier to organize.

Repeat the process for each block of paragraphs until you have a folder like this:

Screenshot of mp3 files
Screenshot of mp3 files

Step 3 - Merge the generated audio blocks 🪡

Now that we have all the audio blocks generated, the only remaining step is to join them. We'll use this free tool:

Click on Add Track and select the mp3 files in numerical order.

Screenshot of, a website to easily join audio tracks.Screenshot of, disable cross-fade

Step 4 - Compress the final file 🐘 ➡️ 🐁

After merging the audio blocks, you may find that the total size of the generated file is quite large.

But don’t worry, to solve this problem, we’ll use this simple tool to compress the audio without losing quality:

Screenshot of FreeConvert MP3 Compressor
Screenshot of FreeConvert MP3 Compressor

Once the compression is complete, click on download.

That’s all 🎉

After completing these 4 steps, your audio file will be ready for publication.

I recommend not deleting the previously generated audio blocks. it's best to keep them in a folder so that if you need to edit the article in the future, you can re-generate only the parts that have been updated, without having to re-generate the entire article.

I hope this guide has been helpful. Thank you for reading!