Deepgram

🔊 Last day at the expo hall for AI Engineer Summit. Be sure to stop by our booth to come say hi before it's too late! We've met so many incredible developers throughout our time here. If you'd like to see what the latest voice AI has to offer, come say hi to our team. See you soon!

💪 Software is becoming multimodal. And these people are helping lead the charge. Here's the update: In this conversation, ‪Vercel's Zeb Hermann and ‪our very own Adam Sypniewski joined Nikita Shamgunov (Neon/‪Databricks‬) to explore how AI agents are transforming the act of coding itself. 🤖 🧠 Questions answered include: What kind of use cases do these AI leaders see multimodal software being most effective? Will multimodal software development be universal, or will it simply be a subcategory of overall computer programming? What currently sucks about coding (with or without multimodal AI) and what's being done to fix it? Check out this video to learn more: https://lnkd.in/gb-xcEFU

📲 Can you tell the difference between a human and AI on the phone? You won't be able to soon. Deepgram CEO Scott Stephenson discusses building voice technology that passes the Turing test while handling millions of calls daily for brands like CVS Health and Anthropic. Listen here: 🍎 On Apple Podcasts: https://lnkd.in/gwSg2Knk 🎧 On Spotify: https://lnkd.in/gXDgp9Xs 🔉 On Amazon: https://lnkd.in/gKkzWnja #Deepgram #VoiceAI #AI #AWS #AWSforSoftware #podcast 

⚠️ New video alert! Our very own Lauren Jo Sypniewski (Head of Data Operations) shares how Deepgram is redefining voice AI by building models that feel human. But here's what may surprise you: She talks not just about scaling data operations and crafting multilingual models, but also tackling real-time *conversational challenges*. But what exactly could those challenges be? 👀 📈 Lauren’s insights on curiosity-driven teamwork and focusing on what matters most will inspire anyone passionate about solving big problems in AI and data. 🤯 Find out here: https://lnkd.in/gC9PxXGS Shoutout to 5X for taking the time to speak with us!

🎤 Voice AI is already automating 25-40% of CVS Pharmacy calls and powering Jack in the Box drive-thrus. Alongside AWS, CEO of Deepgram Scott Stephenson has this to share: Healthcare's desperate staffing needs accelerated adoption faster than anyone expected. The technology is ready and traditional enterprises are transforming – is yours? Listen now → 🍎 On Apple Podcasts: https://lnkd.in/gwSg2Knk 🎧 On Spotify: https://lnkd.in/gXDgp9Xs 🔉 On Amazon: https://lnkd.in/gKkzWnja #Deepgram #VoiceAI #AI #AWS #AWSforSoftware #podcast 

🏆 TMC—a global, integrated media company—has named Deepgram's Aura-2 text-to-speech (TTS) model a 2025 Customer Experience Innovation Award winner, presented by TMC’s CUSTOMER magazine! Here's what's different about Aura-2: "Aura-2 was recognized because it isn’t just another “nice voice” model - it actually shows up for work. Aura-2 is the only TTS model purpose-built for enterprise use cases, rather than entertainment scenarios." - BusinessWire 🚀 Check out the original article in the comments below!

🔥 Our booth at AI Engineer Summit just opened! Come say hello and learn about the latest, most realistic Voice AI models in the world. Meet some incredible people, including: 🧠 Jake Lasky - Senior Applied Engineer 🔊 Jefferson Liu - Staff Technical Product Manager (text-to-speech) 🚀 Simon Crabill - Account Executive ... and many more! We're excited to meet you!

📈 It may be tough to achieve low-latency AI, but it's even more challenging to do so *at scale.* So what's the secret? How do we do it? Instead of scaling hardware, we at Deepgram reengineered the runtime for parallelism and orchestration, achieving stable sub-200ms latency and, in steady-state conditions, around 90ms. 🚀 Details on the engineering are written in the blog below! Check it out 🧠 Blog Title: "How We Took Aura-2’s TTFB from <200ms to 90ms: Engineering Real-Time Voice AI at Scale"

🤯 AI coding agents are getting smarter—and more of them need deserve a voice. That's why we're sponsoring AI Engineer Code Summit in NYC this week! If you're building agents that need to listen, speak, or process audio at scale, come chat with us! What we're bringing: 🎤 Real talk about voice-enabled dev tools 💾 Technical deep dives on Voice AI APIs at scale 🎉 A giveaway you won't want to miss Stop by our booth at the expo. Let's talk about what you're building!

🔊 Thrilled to announce that Brex has announced Deepgram as one of the fastest-growing AI infrastructure companies in the world right now! 📈 Specifically, Deepgram is up 127% month-over-month in spend across Brex customers, outpacing both OpenAI (+51%) and ElevenLabs (+37%) Check out the report in the comments below for further details! 🧠