Deepgram

Nova-3 expands speech-to-text support across Asia-Pacific. 🌏 New support now available for: 🔹 Thai 🔹 Cantonese Traditional 🔹 Mandarin Simplified 🔹 Mandarin Traditional Plus: ✔️ Accuracy improvements for Bengali, Marathi, Tamil, and Telugu ✔️ New Gujarati support Built for production-ready voice AI across streaming and batch workflows. Learn more: deepgram.com/learn/deepgram…

We built Dream Book at AI Hack 2026 ✨ Speak a bedtime story, then watch it become an illustrated, narrated book you can flip through, listen to, and talk with. Huge thanks to @DeepgramAI, @midjourney, and @getsentry for helping bring it to life. devpost.com/software/dream…

We built RocketCursor at the Berkeley AI Hackathon It’s basically Cursor for rocket fluid systems: describe what you want, generate a P&ID, run physics checks, and get failure-aware design feedback. @deepgramai was crucial because rocket design starts as messy human conversation: constraints, tradeoffs, revisions, and “what if” ideas. Voice made that workflow feel natural instead of forcing engineers to type every detail. Check it out: devpost.com/software/rocke…

Here's an updated version of the Farmer's Market Voice Agent demo, using @DeepgramAI. 🚜 Improvements: 🧺 Inventory / purchase tracking 💵 Wallet balance This is kinda fun! 👾 Using the @threejs library. #Deepgram #VoiceAI

Created a simple, interactive farmers market demo for @DeepgramAI Voice Agent. Proximity activates agent. 🎙️🧺🚜 It would be really cool if you could walk up to vendors, in games like @assassinscreed, and talk to them naturally, instead of navigating menus. #Deepgram #VoiceAI

Data + AI Summit is wrapping up. But the Voice Heist is still live! Think you can break our voice agent? 🏆 The challenge sparked some of the best conversations we've had about deploying voice AI in production. Come see us at booth #124 to take on the challenge. 🦸 #DataAISummit

The Deepgram Australia endpoint is now generally available. 🇦🇺 api.au.deepgram.com runs in Sydney, with your audio, transcripts, and speech output processed and stored entirely within Australia. Same API and models. Moving over is a base URL change, no self-hosting required. Blog → deepgram.com/learn/deepgram…

Day 1 at #HLTHEurope, and one thing is clear: healthcare teams aren't asking if they'll use voice AI anymore. 🩺 We're seeing strong interest in how teams move from experimentation to real-world deployment, particularly when it comes to self-hosted deployments, EU endpoints, and meeting regulatory requirements. Looking forward to more conversations and insights throughout the week. 🚀

A chicken combo and a box of donuts don't sound complicated. For voice AI, they are. 🎙️🧩 One lesson from two major restaurant deployments: there is no one-size-fits-all approach to enterprise voice AI. deepgram.com/learn/voice-ag…

Which restaurant order is hardest for AI?

Introducing Batch Diarization V2, a major upgrade to speaker labeling for pre-recorded audio. Understanding what was said is only part of the story. You also need to know who said it. Highlights: 🔹 Preferred 3.3X in human evaluation 🔹 Improved speaker attribution accuracy across real-world audio 🔹 Available today via the new diarize_model parameter Get started: deepgram.com/learn/introduc…

Our Deepgram for Startups program is an excellent way to get additional support as you carry your product to the GTM "finish line". If you are interested, fire an application to deepgram.com/startup-program or reach out to us in our Discord discord.gg/deepgram

📲 Working on UI refinement for the @DeepgramAI Mobile Playground app that I've been prototyping recently. Now we have a hamburger menu to access different core functions. 🍔 #Android @AndroidDev @FlutterDev

See how @premai_io brought real-time voice to its sovereign AI ecosystem with Deepgram Nova-3, delivering low-latency, multilingual speech-to-text that never leaves trusted infrastructure. 🔒 Read the full customer story: deepgram.com/customers/prem…

We're heading to #NiCEWorld26 next week. 🚀 Looking forward to connecting with customers, partners, and innovators shaping the future of #CX. See you there. 👋

😍😍😍 We keep adding on new languages. Keep building and letting us know what we can do to help you build voice into your applications.

If you're building a voice app with @DeepgramAI voice agent, check out our skills repo github.com/deepgram/skills I use it regularly to build demos like for multi language transcription with our Nova 3 model. It's really helped to speed up my dev cycles!

🔊 Use this terminal utility to validate your @DeepgramAI self-hosted stacks running the Aura-2 Text-to-Speech (TTS) model! You can run Deepgram self-hosted on Amazon SageMaker, and using containers (k8s, @Docker, etc.). ➡️ youtube.com/watch?v=mJRjuV… #VoiceAI #Deepgram @awscloud

It was such a pleasure to spend some time at the booth with our friends from @Vapi_AI! Amanda and I had some absolutely banger conversations about developer communities - stay tuned for some really cool stuff both in Vapi's spaces and in our spaces~ discord.gg/deepgram

@Vapi_AI @hey_amandam @NaomiLGBT @CascadiaJS We wouldn't have it any other way~ 🩷

@hey_amandam @DeepgramAI @NaomiLGBT @CascadiaJS Vapi 🤝 Deepgram

Shout out to @DeepgramAI community manager @NaomiLGBT for being clutch offering her hotspot during my workshop at @CascadiaJS to literally explaining the @Vapi_AI platform at our booth. Vapi partners are the best! Definitely check out the community she's building over at Deepgram

The improved Nova-3 Portuguese model is here! 🇵🇹 🇧🇷 We’ve enhanced the Nova-3 Portuguese model with improved transcription accuracy across Portuguese language variants, including Brazilian Portuguese and European Portuguese. Available now with: model="nova-3" Supported language codes: pt, pt-BR, pt-PT Try it in the Playground or get started in the docs: playground.deepgram.com/?endpoint=list…