Hello sir. Hello sir. How's it going? Good to see you, man.
Very excited. Very excited to see you coming in. Thanks for having me here.
This is great. You have a good setup. What do you think?
No, it's a great space. It's pretty cool, right? No, it's great. You have good energy, good immersion into the voice space. So that feels on topic for us.
Yeah, absolutely. I love the jacket, by the way. We've got to try and get a hold of some of those sometimes.
Do you rock that every day? Like, I was curious when we saw you — where did we see you? In Paris? We saw each other in Paris. Yeah, and I was like, I've literally never seen him not wear one of those jackets.
I only wear these these days. It's this, or we have those quarter zips. So when we are deep into the night, I suddenly swap for the quarter zip.
I have a quarter zip. Okay, great. Did you bring it over? No, I was at — I think it might have been pre-ElevenLabs for a launch that you guys did.
Yeah, maybe like a year and a half ago. Dawn Capital's offices.
Yeah, yeah, yeah. If you remember the V3 or V2 — maybe it was V2, I think — but I got on there. And I have a cap as well, which is great.
But we have something for you. Actually, Monty — Monty, whilst we were away, when Amjad came in, we got him a present because we're fans. But what it is from Monte, I'll go a little closer. So, we know how important your high school was to you in Warsaw. You met your co-founder.
So, we got you — we built you a...
The logo and then your class photo.
No way. Amazing. Thank you, dude.
I hope you like it. It's just a small memento to take away from the ElevenLabs studio.
This is amazing. You know, it's crazy, but this — so the picture is taken when we were class 2. And the 2 is now our logo, or it's the logo effectively.
Oh, dude, that looks super familiar. It's kept in the picture. Let's go. I love that. I love that.
Apart from that, there are a few other people from 11s here. So we have Magda here. So it's quite a few people that joined ElevenLabs' journey since.
You must be like one of the only companies that has, you know, that many employees from their high school still working.
Yeah, just on the side. It's great.
That's great. Legend.
That's insane. And we have to talk about Eleven v4. Congrats on the launch. Do you want to give us a rundown of exactly what it is? You guys have crushed benchmarks — we were just going through it — but would love to hear from you.
No, thanks for having me here, first of all, and I'm very excited for what you guys are doing for the European ecosystem and for what we hope to do for the European ecosystem over the many years to come. But ElevenLabs started in 2022, so almost five years ago, and the first thing it started with was — and as a company we combine research and product. The first model was a text-to-speech model, the first model to finally move past the uncanny valley of speech, where the quality of turning text to audio sounded as good as the human.
And that was the first one for the industry. It suddenly was good enough for narration, for voiceover. And this is a full-circle moment since that moment, because now, five years later almost, we release a new version of that model that covers now 90 languages. So it can do extremely high quality across Japanese, Mandarin, Hindi, all the variants in other languages and dialects in those — it can do Catalan. So it's really a global model.
And once again, it's leading on all benchmarks. So it's a model which is crazy to see that over all those years, my co-founder Piotr and the research team behind are time and time again being able to outcompete the big and the small companies alike across that modality. And that happened once again with the v4 launch with the text-to-speech model, over the years that expanded to speech-to-text, to orchestration and TTS models for voice agents, to localization models, to music models. But being able to ship that model yesterday was a big milestone, and the model is coming with an additional set of functionality over the classic text-to-speech.
Super high expressivity. So you can finally generate all the non-verbal elements that are tricky, whether it's laughter, whether it's the whisper, whether it's almost the singing part. It's extremely reliable, so you have great consistency across long fragments. It supports the 90 languages. Then we have a version optimized for conversation, where you actually can understand the emotions on the other side and pronounce that in a much better way. And you've probably seen some of the benchmarks, but on the benchmarks it jumps. It's another step. So just proud of the research team and all the people that are just crushing it. And it's hard. It's hard because you've probably seen the number of people building in the voice and interaction space today.
Every company is building in that space. So being able to ship that yesterday and show what's possible has been tremendous once more, and now it's shipping across all of our products, whether it's our agents, whether it's the creative space, whether it's the wider set of API users. We are bringing it everywhere.
That's amazing. Let's just quick clap for... We got... I'm clapping for the v4 and the team back there.
Shout out to the team as well. Shout out to the team. If we could just zoom out a little bit on the growth trajectory, I'd love to know from your position, how has it been being at the epicenter of this? Obviously you have an amazing team around you. When we spoke to Luke ages ago, he mentioned that the team still operates with small SWAT teams and everyone's horizontal. There's no vertical hierarchy. Everyone's just moving at startup speed 24/7, and you see it reflected in the ARR numbers. How fast are things growing internally from the monetary side, from the value side? Because I imagine that the ACV deals you're doing on the enterprise side now are just getting bigger and bigger.
