Elevenlabs, which is a hot AI sound start, $ 3.3B confirms CSAMING SERION C ASSING in an assessment

OnelabsOne of the more popular beginnings working in the field of AI, on Thursday, the company was ROUR ROUR ROUR, 180 million CM ROR, to estimate the value of $ 3.3 billion. The growth of A16Z and Iconiq is an international investment.

Fandrays’ was rumored First reported by TechCrunch. The last numbers confirm a little, but not all of the details we previously reported (especially the total size of the tour; evaluation and lead investors are the same).

Financing will be used to continue to build Elevenlabs for audio tools and business development.

Piotr Dabkowski with Piotr Dabkowski with Piotr Dabkowski, the Qur’an Mati Staniszewski said that the beginning was investigating the establishment of AI models that further control the further expressive and more control. Staniszewski also focuses on “Omni-Models” in the company’s “Omni-Models”: Multimodal connects text-based models with sound models.

There was anger of investor interest in Onelabs A few months backbehind two important streams. First, the generative AI holding a lot of companies in the awakening has been a large wave of hype around the AI. Second, Elevenlabs appeared as a great player among those who provide synthetic sound technology. Dozens of core publishing and content as media and play, as well as a number of other technological beginnings, all use eleventabs technology to strengthen sound and sound features.

Not surprisingly, this has become a very crowded tour with very prominent names.

The new investors in this series include Nea, World Innovation Laboratory (WIL), brave, Catalyst Catalyst Foundation and Abu Dhabi Investment Firm Lunate. Sequoia Capital, Salesforce Ventures, Smash Capital, SV Angel, NFDG and extensive lighting included in the former investors.

In addition, Onelabs are also a number of new strategic supporters, ie companies that use the technology that is now investing. They include Deutsche Telekom, LG Technology Ventures, HubSpot Ventures, NTT Docomo Venesures and RingCentral.

Seth Pierrepont, A16Z and company co-founders Jennifer Lee, will join the company.

Iconiq has expanded its activities around the generative AI beginnings. By striking the written output, the company A round of $ 200 million in the writer last November.

“We’ve always felt that the sound is always a very important modality, and we thought it would be a very big company in this category,” said Pierrepont Techcrunch. “We observed Oneevenlab, and how quickly the technology rose in terms of quality, mind and acceleration, and how quickly the founders rose in terms of domain experience.”

PierrePont added that as a member of the Management Board, there will be many conversations with the company, creating new usage cases for sound and find the necessary markets for it.

At a time when the beginnings are still difficult to close growth tour, he raised $ 80 million series of $ 80 million worth of $ 8 billion, a year ago. Elevenlablar gathered a total of $ 281 million so far.

Product road map

In addition to a focus on improving AI models, the company plans to use the financing to develop a bidder aI builder to reach more consumers through direct and partnership.

The company debuted last year AI Talk Agent PlatformAnd the main part of this product has developed a speech-text component. Staniszewski noted that the company wants to develop more in that area.

“We want to understand what a better conversation is in your conversation. We are working on ways to move away from the content and understanding and transcription,” said Staniszewski. “Many people say that the text is a solution to the text. But we think it is quite bad. We think we can build better speech models because we can build our teams and give us quick feedback.”

The company also wants to reduce spoken agents working in AI by supporting its inherited communications such as telephony and better integrate various sources of knowledge. This is partly a partner in this round with TELKOS.

It is also used to touch their own archives of customers. Onelbabs partnered last year Publish time to place a spoken bot To ask users about the second of the year.

Staniszewski provides more conversing AI agents in the company’s websites: news sites, for example, users will be able to ask questions about stories or ask them to summarize them.

CEO also noted that although the quality of the EU-power sound boots improves, and will not sound natural while responding to people or reacting to people in different ways.

“The way I talk to you how you react to me or I will be anxious, or sometimes I will defeat you. That’s why the current gene AI solutions are good, but people are not as good.” Said Staniszewski.

Iconiq’s Pierrepont, stressed that when speaking AI, the machine communication decreases, and users are immediately losing interest.

Elevenlabs has mainly raised (and income funnel) through the B2B partnership (and income). But it also comes out directly on a muscle.

In 2024, the first purely consumer-looking product, Elevenlabs reader, An application that reads articles, text and documents. Later added company capability Create a podcast with generative AI sounds From the documents and web pages – not different from Google notbooklm. Staniszewski said he wanted to expand his consumer experience.

Can actually do that. TechCrunch saw that the company’s audioobooks did not test a program, broadcast on the audiobooks on the company’s platform. The company also wants to provide tools to read a voice book to read more than one voice read more than one voice read when creating better localization.

Staniszewski noted that the company’s users and companies have been identified in their application, including better distributions in their application. There will be something to keep track of their customers to be brought to a true direct competition. (This has been a reason for many B2B technological companies to prefer to be away from consumer plays. LIGHTSPEED-supported Pocket FM and Google Supported Cuckoo FM.

Elevenlabs gives the Ai created audio in products such as portability, rabbit, R1, Chess.com, ESPN, Lex Friedman Podcast, Atlantic and Settanism. The purpose for the company has the last conversation stack until the end to be able to be in more places and can create more experience and understanding for customers.

Safe

All Elevenlab was not silver, not cloudless: his technology was applied in several remarkable wrong information campaigns. The last report of the threat to the intelligence company noted the future The company’s product was used in Russian propaganda operation. Used the company’s voice platform to create someone last year Joe Biden’s Audio Depth. In 2023 Hakharding Joe Rogan, Ben Shapiro and Emma Watson, allegedly used by the 4Chan member’s AI sound generation vehicle, used the sounds to create sounds to spread the problematic content.

But the company replied quickly. Today, there is a policy that prohibits “unauthorized, harmful or deceptive.” Moreover, it uses human moderation that leads the machine to weed, such content. However, the company also has a set of tools and has another direct consumer touch points, which opens the door to more opportunities for the abuse of harmful actors.

“As one of the Frontrics of the AI ​​sound work, we treat the correct security mechanism when building technology.

Staniszewski added that when the company pursues C2PA, it is a standard to track the content using metadata, which social vehicle This allows everyone to test the use of Elevenlabs technology using digital signatures in votes during the generation. In this case, it can be a way of developing approaches to the development of development.