DeepL, a global leader in language-based AI, announces that the team and technology behind Mixhalo, the San Francisco-based platform for real-time, ultra-low latency audio, is now part of DeepL.
The investment strengthens DeepL Voice, the company's AI-based real-time voice translation solution, and expands its use cases to larger and more complex environments where speed, clarity and reliability are critical, such as major events, conferences, customer support and business-critical workflows.
DeepL Voice is already a leading solution for real-time translation of spoken conversations, from virtual meetings to in-person conversations. According to an independent evaluation by Slator 2026, the solution achieved a quality score of 96.4 out of 100 and an error rate of only 4 percent, compared to the market average of 17 percent. According to the report, it outperforms alternatives such as Microsoft Teams, Zoom and Google Meet in terms of accuracy, fluency and reliability.
By integrating Mixhalo's team and technology into DeepL Voice The platform now supports ultra-low latency audio distribution in large-scale live environments. This means that both translated speech and captions can be delivered clearly and instantly to audiences ranging from small groups to tens of thousands of participants, while maintaining the pace and natural flow of live communication.
– DeepL Voice is already changing how people and businesses communicate across language boundaries every day, and Mixhalo’s team and technology enable us to take this to even larger and more complex environments, says Jarek Kutylowski, founder and CEO of DeepL.
“The team has solved one of the most challenging problems in live audio: delivering high-quality audio to thousands of people simultaneously with virtually zero latency. Together, we are building the AI layer that enables real-time language, so people can understand each other naturally wherever they meet, whether it’s team meetings, client calls, or large international events.”.
Mixhalo founded in 2016 by Incubus guitarist and songwriter Mike Einziger, violinist Ann Marie Simpson-Einziger, and tech entrepreneur Vik Singh. The platform was developed to change how people experience live audio by delivering high-quality synchronized audio to thousands of listeners simultaneously with extremely low latency.
The technology has been used at major sporting events, entertainment productions and brand activations for organizations such as MLB, NASCAR, Equinox, Verizon, T-Mobile, Sacramento Kings, Red Rocks Park and Amphitheatre, and TSX.
Mixhalo has also been used at leading international conferences and events such as CES, Mobile World Congress, Salesforce flagship events, Databricks AI Summit and Microsoft AI Tour Paris, as well as concerts and residencies for artists such as Metallica, Aerosmith, Sting and Charlie Puth.
– We created Mixhalo to deliver the highest quality audio to audiences of all sizes with a latency of less than 20 milliseconds. When we started working on real-time translation, the biggest challenge was finding technology that could keep up with live speech without compromising quality, says Vik Singh, co-founder of Mixhalo.
“After testing solutions from virtually every leading vendor, DeepL was the only one that could deliver accurate translations at the speed required to create a natural conversational flow. We have already seen what this combination can achieve in real-world environments, and by becoming part of DeepL, we can now offer the experience to even more customers and audiences around the world.”.
DeepL and Mixhalo have already launched the joint technology, using the DeepL Voice API to enable real-time translation via Mixhalo's live audio platform. The teams are also exploring use cases in customer support through integrations with Amazon Connect, among others.
– Voice-based AI is ultimately about latency, quality and reliability in real-world situations, says Sebastian Enderlein, CTO at DeepL.
“The Mixhalo team has extensive experience implementing APIs and audio infrastructure in environments where there is no room for delays or errors. That expertise will be invaluable as we continue to scale.” DeepL Voice, while their presence in San Francisco brings us closer to customers, the partners and developer ecosystem that are shaping the next generation of AI products.
The announcement also marks the opening of DeepL's first office in San Francisco, strengthening the company's presence in the US, its fastest-growing market. DeepL's American customers already include NVIDIA, Cisco and Nasdaq, and almost half of all Fortune 500 companies today use DeepL's solutions.
Visitors will be able to experience the technology in action at GITEX Europe 2026 in late June, where DeepL is the official translation partner and will deliver live German-to-English translation on the main stage. The company will also demonstrate its next-generation Voice AI capabilities at several upcoming events in the US, including Databricks Data + AI Summit, Esri User Conference and Salesforce Dreamforce.
About DeepL
DeepL is a global AI company building the language infrastructure that powers global business. More than 200,000 business teams and millions of individuals use DeepL's language-based AI platform to communicate globally, collaborate, and work across languages in real time. By combining cutting-edge AI models with enterprise-grade security and privacy, DeepL enables organizations to work seamlessly across markets and cultures. DeepL was founded in 2017 by CEO Jarek Kutylowski and now has more than 900 employees and is backed by leading investors, including Benchmark, IVP and Index Ventures.
Read more at www.deepl.com.
Swedish
Danish
Norwegian
Finnish
English
German
Dutch