Deepgram

The wait is finally over, and we're thrilled to present Aura — the first text-to-speech model built for responsive, conversational AI agents and apps. 🗣️ Read the full announcement: https://lnkd.in/g4ra5vMq Over the last year, we've heard our customers' heartache about the current crop of text-to-speech products, citing roadblocks related to speed, cost, reliability, and conversational quality. “Deepgram and Groq share the belief that speed and efficiency are the missing ingredients in unlocking natural AI for daily use by everyone, as evidenced by the recent viral reception to ultra-fast LLMs when made available for the first time. Their voice AI models are prime examples of what can be achieved with the Groq API.” –Jonathan Ross, CEO & Founder of Groq That's where Aura comes in. Designed to handle real-time conversations at scale, developers can create realistic AI agents to support seamless interactions across various cases, from voice ordering systems to customer support. Aura checks all of the boxes: ✅ Lightning-fast speed with less than 250 ms latency ✅ High-quality voices with natural-sounding tone, rhythm, and emotion ✅ Cost-efficient for high-throughput applications Check out our open-source interactive demo: https://lnkd.in/gMJTDWE8 We're eager to see how Aura will fuel the next wave of AI innovation, and can't wait to see what you build!

We had a great turnout of leaders at our AI Dinner last month in Palo Alto! We are always inspired by the innovative minds shaping the future of AI. We look forward to driving voice technology evolution together!

The last day at the expo hall at AI Engineer World's Fair! Stop by the booth to learn how developers build with our APIs to power use cases ranging from AI agents to live captioning. Also, join Scott Stephenson, at the breakout stage today to explore the principles and best practices for creating responsive and realistic AI agents! 📅 Live today at 2:20pm PDT!

The world’s fastest voice bot was built using Daily, Cerebrium, Deepgram, and LLaMA. It’s open source, and this article tells you exactly how it works: https://lnkd.in/gFrMNyaG

From a packed workshop to a busy booth at AI Engineer World Fair! Stop by our booth for the welcome reception to learn how developers use our latest voice AI models to boost productivity and build scalable AI agents 🚀 #aiewf #aiengineerworldsfair

#CCW is a wrap! 🌟 We had a blast meeting so many CX leaders and showcasing how voice AI unlocks customer insights at scale. A special thank you to everyone who attended our AI Mixer with Twilio at the Brooklyn Bowl! Dive deeper into our contact center solutions or reach out to learn more. https://lnkd.in/gQmEqCQ6

Join Deepgram at the AI Engineer World Fair next week! We'll be showcasing how to integrate voice AI capabilities into your products for a wide-range of use cases from AI agents to live captioning. Plus, catch our team on stage to hear: ▶ How to build and scale AI agents to handle 1,000+ concurrent conversations ▶ Best practices for designing multimodal interactions If you're attending, stop by booth #S5 to learn more! #aiengineerworldfair

Last week our team attended #CCW Vegas to showcase what's possible with voice AI and share our latest Five9 Studio 7 integration with the CX community! Five9 Studio 7 users now have access to Deepgram's powerful transcription to build high performing voice bots at scale! 🚀

We are thrilled to unveil our latest innovation tailored to healthcare use cases, the new and improved Nova-2 Medical model! ⚕️ The medical industry poses many challenges including complex terminology, noisy environments, diverse speakers and accents, and strict data policies. With this release, we extend support to our wide range of customers including Lyrebird, Augmedix, Tortus, Middletown Medical, Twilio, Five9, and Phonely, who are driving healthcare innovation by advancing medical documentation and patient interactions with voice AI. Building upon our industry-leading model Nova-2, this model addresses common challenges within the medical domain and delivers 20% higher accuracy on medical terminology compared to leading alternatives. The Nova-2 Medical model is capable of transcribing symptoms, diagnoses, treatments, medications, and clinical jargon, making it possible to automate EHR and SOAP notes, patient intake, and more. TL;DR: this latest model delivers: 🎯 Superior overall accuracy: an average 43% relative word error rate (WER) improvement vs. benchmarked alternatives with 11% greater accuracy than the previous model 🩺 Enhanced recognition of medical terms: 20% higher word recall rates (WRR) on average vs. leading competitors; 16% relative WRR improvement vs. previous model Review the benchmarks in the full announcement and learn how to get access to the Nova-2 Medical model!👇 https://lnkd.in/gUnsJ4pq

We are excited to announce that Deepgram has strategically acquihired Poised, the innovative team and proprietary technology behind Poised AI Coach service. The Poised AI Coach is an advanced communication tool that leverages Deepgram’s voice AI to provide real-time, actionable feedback during online meetings.   With this acquisition, Deepgram continues to build on its industry-leading AI models for speech to text, text to speech, and audio intelligence. The addition of Poised AI Coach will not only showcase the future potential of voice AI, but also provide immediate, substantial value to users seeking to improve their virtual meeting experiences. Read the full announcement here: https://lnkd.in/g-VyJEhx