Text to Speech is the foundation of every voice product you build. Audiobooks, agents, dubbing, accessibility, games, IVR. They all start with turning text into audio that sounds human. Here's how it works in ElevenLabs. Made with ElevenLabs & Hyperframes
Tadas Petra
@tadaspetra.com
Developer Relations Engineer @elevenlabs.io Founder @hungrimind.com
I wanted to test how good the Voice Changer API really is...
Voice Changer transforms any speech into a different voice while preserving the original timing, emotion, and delivery. Send an audio file to our API and get back the same performance in a new voice, ready for character work, localization, or fixing lines in post.
Voice Changer transforms any speech into a different voice while preserving the original timing, emotion, and delivery. Send an audio file to our API and get back the same performance in a new voice, ready for character work, localization, or fixing lines in post.
Voice Isolator removes background noise, music, and ambient sound from any recording. Send an audio file to our API and get back a clean speech track, ready for transcription, dubbing, or post-production.
We built a Lie Detector AI Agent. It listens to conversations and interrupts when it detects a lie: - A loud buzz - An angry pirate calls it out in real time Built with ElevenLabs Agents. youtu.be/SMp1M1LmYwE
We built a Fact Checker Agent
YouTube video by ElevenLabs Developers
youtu.be
Introducing ElevenLabs Devs, a new YouTube channel for AI engineers. Expect deep dives, demos, and clear explanations of key concepts across Text to Speech, Speech to Text, ElevenAgents, and broader AI systems. Subscribe: www.youtube.com/@elevenlabsd...
What are Commit Strategies? When you stream audio with ElevenLabs, you get: - Partial transcripts - live interim text. - Committed transcripts - finalized segments after a commit. Commits define structure, latency, and downstream behavior.
How to Publish your own Agent Skills - Create a repo - Add a SKILL .md file into your repo or a multiple folder structure - Add a README .md file to explain your skills - Install using npx skills add <owner>/<repo-name> - Wait ~1-2 days to get crawled and listed on skills .sh
I finally tried ElevenAgents. Using Skills, I integrated an agent into my Flutter app in under 15 minutes. As a demo, I can call the agent to help strategize YouTube videos. It generates the title and description directly inside the app.
Learn how Scribe v2 Realtime works. In this overview, we explain how our ultra low latency Speech to Text model streams audio, generates partial transcripts, commits accurate segments, and enables real-time voice applications. elevenlabs.io/blog/how-sc...
ElevenLabs
elevenlabs.io
The best new features in Flutter 3.38 - Hot reload on Web Server (this will speed up @hungrimind a ton) - Web development configurations - Better Widget Previews - Performance and bug fixes
The hardest thing for me to accept about entrepreneurship is that hard work does not result directly in success.
My fine-tune model failed due to "hate, self-harm, sexual" moderation checks. But the training data was about teaching how to code. GPT said that it's due to snippets of a word. There's no way, right?
After a year of working full-time on @hungrimind, we finally have a landing page.
As an employee, I thought mission statements were cringey and stupid. As a founder, I realized they are the most important part of a business.
Does anybody else find macOS 26 feels kinda childish? - the corners are too round - the buttons are way bigger - usability not better It feels like this design was made to be more dopamine-inducing than productive
It's wayyyyy easier to refactor a working product than to make a well architected product work.
We are adding a roadmap to our changelog so you can see what we are working on. What do you like better? 1️⃣ Kanban 2️⃣ Only the In Progress (less clutter) Also surprise! We are working on a Dart course!
We added a changelog, and I think it's the nicest changelog I've ever seen. What do you think? www.hungrimind.com/changelog
Pretty crazy that we went from a computer filling a room to a thin little device. And fundamentally, they both work the same way. Little transistors turning ON and OFF.
I'm calling this Sep-Dec season "The Great Lock In of Q4 2025". If all goes well, history books will write about this period.
The first weekend of the Sep-Dec Lock In was an overhaul of the site to have a cohesive brand and professional look and feel. Now the entire site follows the design style I call: "Modern Analog". I couldn't be prouder. It feels like the site is becoming an extension of us.
In 1 generation, we went from my dad having to work night shift after school since he was 8, to me building a worldwide educational platform full-time I worked my ass off to get here, but I owe my parents everything because they got us 0 to 1. I just kept it going
finally updated our logo to match the stroke of the text, ready for the MRR inflection point now
I feel like I've been lost in the sauce for the past year, and never got to fully appreciate what an absolute pleasure it is to be doing what I am doing on a daily basis.