Mirelo Bets Big on Fixing AI Video’s Sound Identity

0:00A company called Morello has just raised$41 million from Index and A16Z to solve one of AI videos biggest problems and that is not having sound effects background music or audio in the background of ai generated video clips today on the podcast we'll be diving into the state of ai video generators and what this company morello plans on doing with their 41 million dollars in this seed round that they have just raised who gave it to them what the plan is and where the market goes in the future before we get into all of that if you want to try all of the ai models i talk about on the show, make sure to go over to AIbox.ai.

0:40There's over 40 different AI models from OpenAI to Gemini to Claude to Grok to 11 Labs, 40 other ones. And instead of paying subscriptions to all of these different companies, you can pay$20 a month and get access to all of it in one place without having to switch tabs, have multiple subscriptions. I hope this saves you a ton of time. You can also compare models side by side. Go check it out at AIbox.ai. All right, let's get into the show. This is an interesting company, and I also think that it's an interesting problem. And while I'm not saying that Morello is, you know, going to be completely struggling as a company, they do have a lot of headwinds.

1:16So they have just raised$41 million. This is their seed round. It was led by A16Z and Index Ventures. So obviously, they're getting a lot of VC funding that's going into it. Earlier this year, this is a company that was based in Berlin, but they just released SFX v1.1. So essentially, this is an AI model that it watches a video you upload to it, and then it's going to add synced sound effects to the model. To be honest, I think this is something that's really impressive, but the headwinds we'll be facing are from Google and a lot of other AI labs that are already including sound effects into their videos that are being generated.

1:51Now, this wasn't the case when this company first started out, but today that is the case. So the new money that they've raised, they said is that it's going to help them to compete against Google and other people. They've been in stealth mode for a very long time. And when they were in stealth mode, they didn't have a ton of resources. There was a bunch of other companies, Sony, Tencent, they released video to sound effect models. Also, Kuei Shu, who is owned by Kling AI out of China and Eleven Labs, they've all released models that do the same thing. And so while Morello is different from them because it's a lot narrower of a focus.

2:26A lot of those other companies, they're doing more of like the sound and the audio. And right now Morello is just focused on the sound effects. So I saw a demo video where it's like a girl walking in, you know, the snow and you can kind of hear like the wind blowing around her and the crunching of the snow under her feet. And so, and it was, you know, it lined up with her footsteps in the video. So you could hear the crunch of the snow as her feet made a step. And to be honest, it was like, this is, is a great tool. This is a great value add. I see this as being very valuable. But the question is, is this not going to be something that a lot of these other video generators have baked in?

3:00Now, I think the argument for Morello really succeeding is if AI companies can't figure out how to do that, and they use the API of Morello and just plug it straight in if the video companies do that. But I tend to think that they are they're going to get there. According to the CEO, they expect their startup to double or triple their headcount right now they have a team of only 10 people. So raising$40 million with a team of 10 is very impressive. Their CEO is CJ Simon Gabriel. And they've been working on this out of Berlin, as I mentioned, all of the new hires are going to support their R &D.

3:33They also have product and go to market strategies that they're working on. And they've recently published their models on fal.ai and also replicate. And they expect the API usage to drive most of the revenue in the short term. So this isn't something that they're expecting users to go and use a lot of they're really expecting companies that are interfacing with AI video generators to kind of plug these two things together. And now I will, like I know earlier, I said, I expect a lot of AI video generating companies do this themselves. But in the meantime, I mean, you have to, like, I think I have to give them their flowers in a sense that there are a lot of video generators that don't have sound.

4:10And video editing platforms, I think can make really great use of this. In addition, I think, you know, a lot of people will say that Google is going to basically beat them because Google can already do this. They have the sound effects and the video combined and kind of the Gemini Google generators, video generators that are out right now. But I think there is a lot of use case, if you can imagine, built into something like Adobe Premiere, where maybe you have, you know, footage that was shot. Maybe there just is no sound. Maybe you grabbed copyright free footage off online that didn't have any audio to it and you want to add something after the fact, right?

4:45You don't want to go and like kind of regenerate a whole video for the clip that you have. You just want to add sound or change the sound in it. And so I think that there's a lot of, or maybe, you know, your microphone is off or your microphone was bad on your camera, whatever. I think there's a lot of cases where you have footage and you want to add, uh, you want to add audio to it. And so I think it's probably less of like a creator by creator using their platform for it explicitly, but I think they're right. the API usage. So having their software plugged into other software will be a good solution for this.

5:14They said that they're investing in building out their workspace for creators. They have something called the Morello Studio, which is going to eventually support kind of like a full professional use software studio. So I think this is interesting. I also think plugging it into other products is a really good way to go. So as they're preparing to scale right now, they're also anticipating concerns around training data that have been difficult for a lot of other AI generated companies according to Georgia Stevenson who led indexes investment into this company they they base their model on public and purchase sound libraries and they're signing revenue sharing partnerships that are giving money back to artists who have rights to some of those some of those sound effects so you know they'll go and they'll pay for them and then they're they have some sort of royalty program.

6:01So, you know, I think this is a good way to go. It's definitely something that's tricky with a lot of these AI generated tools right now. I think that, you know, people are concerned they're going to displace musicians or sound designers. They currently have a freemium model that has a, like, and also a$20 a month plan. And so they said that they're mostly targeting amateurs and prosumers that are wanting to unmute their AI generated videos. But I think you could also see a pretty great use case for regular videos getting these sound effects as well. When their CEO was asked about where they kind of stand in the market, there's a funny quote where he said, George Lucas said that sound is 50 % of the movie going experience.

6:42It's not an overstatement. If anything, it's an understatement. You can take exactly the same image and the sound will shape a completely different ambiance depending on the sound and the music that you put in there. So he has a co-founder as well who's named Florian Wenzel. They're both AI researchers and they're also musicians. So they said that AI music generation is on their roadmap, but I think they are seeing more pull from the sound effects. They obviously are more integrated there. I think part of that is because less researching is happening there than in other fields. And so their company stands out.

7:13They said, quote, it's easier to build a real moat here and then capitalize on it. So I think this could pay off for Morello. their valuation has increased very significantly since they did their pre-seed round that was led by Atlantic which is based in Berlin as well they also have some participation and in this new fund that or this new round that was just raised so total they've raised 44 million dollars which is you know pretty fantastic for for the whole company they are also backed by some angels who give them a bunch of credibility. They have the Mistral CEO, which is Arthur Mensch, and also the Hugging Faces chief science officer, Thomas Wolfe.

7:54They also have the fal.ai co-founder, Birke Gur, a bunch of other people. So the team is obviously trying to work quickly. We know that these AI generated videos are not going to probably mute forever. And so I think they want to get their technology as advanced as quickly as possible. Google Gemini has their video generator that now incorporates soundtracks, which is powered by DeepMind's VO 3.1 audio to our video to audio model. So if anything, I think that they know time is of the absence and they're trying to really quickly move on this on this technology. In any case, if this was an interesting podcast to you, make sure to leave a rating or review wherever you get your podcasts.

8:32Thank you so much for tuning in. I really appreciate you stopping by. Make sure to check out AIbox.ai if you want to try all of the latest AI models in one place and I will catch you in the next episode.