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…

A Fortune 50 retail pharmacy scaled its Interactive Voice Response (IVR) to 1M+ calls per day across 7,000+ locations with Deepgram. 🏥 📈 92% recognition accuracy on pharmacy vocabulary 🧠 85.3% on complex medical terms 📞 14,000 calls/hour in production ✅ Higher containment, fewer misroutes, lower agent costs. They replaced a legacy Nuance IVR with Nova Medical (STT) and Aura (TTS). The architecture supports multilingual expansion and new workflows without replatforming. Full story: deepgram.com/customers/fort…

Your AI voice agent survives production if it can handle…

@BaymuratAbdumu1 🔥🔥🔥

@DeepgramAI 🔥

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…

@shubhankar8757 @cartesia @msft4startups @Azure Glad you find it valuable! I hope we'll see you in the Slack!

The startup grants by providers is the most goated and under appreciated thing I have ever seen. Thank you @cartesia @DeepgramAI and @msft4startups @Azure

@deepgramscott Should I also wear a helmet during flow state? 🤔

A little about me: I’ve always loved motocross. You can’t debate a jump. You either commit or you don’t. If you hesitate halfway through, it usually ends badly. Most of what I do day to day rewards more thinking. On a track, that doesn’t exist. The feedback loop is immediate and very clear. I love the thrill, and it’s a great reset day for me.

Today we're announcing Flux Multilingual for Restaurants. 🌎 🍽️ In a diversifying world, brands that force customers to order in a language that isn’t theirs risk lower customer satisfaction, weaker brand trust and slower service. Restaurants can now deliver faster, more accurate multilingual interactions across ordering, support, and service experiences. Available in English, Spanish, French, German, Hindi, Russian, Portuguese, Japanese, Italian, and Dutch. Learn more: deepgram.com/learn/your-res…

Real-time voice: there are huge challenges, but they're not impossible to solve. Here's @deepgramscott, at Cerebral Valley Voice:

@madvortex_ Glad to hear you're as excited about it as we are! We would love to hear about the cool things you build with our MCP!

@DeepgramAI the mcp integration is what makes this interesting. voice context loaded on install means it stops being a standalone tool and starts being part of how the agent actually thinks.

AI coding tools still struggle with speech APIs. So we shipped 3 tools in April to fix that: 1️⃣ dg CLI for terminal speech AI workflows 2️⃣ dg mcp for AI coding tools 3️⃣ deepgram/skills so your agent loads Deepgram context on install Now your agent can build with Deepgram context already loaded. The eval-to-prototype loop just got a lot shorter. Blog: bit.ly/48Oz2om CLI: bit.ly/4tYKiqQ Skills: bit.ly/3QY5jDg

Big congrats to @Vapi_AI on their $50M Series B at a $500M valuation! Vapi makes it easy to build voice agents with people skills. ✅ 2.5M+ agents launched and counting ✅ 750K+ developers from startups to Fortune 500 ✅ 1 Billion calls All powered by Deepgram's leading voice APIs. "Our voice capabilities powered by Deepgram APIs deliver the accuracy, low latency, and quality our customers need to build reliable, real-time voice agents." — @nikhilro_, Founder and CTO at Vapi #PoweredbyDeepgram

Deepgram @ #SIGNALConf We’re onsite in San Francisco today! While you won't find us at a booth, we’re bringing the latest breakthroughs in Voice AI to the stage. Whether you're looking to eliminate latency or build sophisticated, human-like agents, these are the sessions to be in: 📍 11:00 AM | Meet Flux: Build Real-Time Global Voice Agents with One Model | Lighting Talk 📍 11:40 AM | Zero Latency, Total Precision | 200 - Intermediate 📍 2 PM | From Bots to Best-in-Class | 300 - Advanced Our team will be hanging out after each talk—come find us to chat about the future of Conversational AI or just to say hi. See you there!

Deepgram STT is now natively available on @togethercompute. One platform, full voice-agent stack: Deepgram transcription, Together-hosted LLMs, Aura-2 TTS. Sub-3-second round trip on reference builds. If you're prototyping a voice agent and bouncing between four vendors, you can stop. Build your voice agent on the Together x Deepgram stack: deepgram.com/learn/together….

🔊Tomorrow: Deepgram’s TTS product and research teams walk through how to build a TTS evaluation pipeline for production voice agents. Scoring rubrics, automated metrics, structured human eval. 10am PT / 1pm ET. Register Now: deepgram.com/webinars/how-t…

🌐 “Multilingual” shouldn’t mean multi-system. Yet most real-time speech stacks still look like this: → language detection service → per-language models → routing logic → latency you can’t explain There’s a better way. 📈 Deepgram’s Flux Multilingual runs 10 languages in a single real-time stream — with: ✅ automatic language detection ✅ native code-switching ✅ language_hint biasing ✅ one model, one WebSocket No routing layer. No model hopping. No duct tape. If you’re building voice agents, this changes your architecture. Read the technical deep dive ↓ deepgram.com/learn/flux-mul…

Full-stack Voice AI in one pipeline? Yes—here’s what it actually looks like ⚙️🔥 This deep dive by @asifrazzaq1988 shows how to build an end-to-end workflow with Deepgram’s Python SDK: real-time transcription, async audio processing, TTS generation, and text intelligence (sentiment, topics, intents). We’re talking word-level timestamps, diarization, summaries, keyword boosting, and scalable async execution. If you’re building Voice AI systems, don’t skip this 👇 marktechpost.com/2026/04/24/a-c…

@Cloudflare Workers AI's speech layer runs Deepgram Nova-3, Aura-2, and Flux at 300+ edge locations worldwide. For developers, that means voice agents that answer in the time it takes a caller to finish breathing in. Cloudflare handles TLS, DDoS, and caching; Deepgram handles the speech. Cloudflare is the platform, the voice models are powered by Deepgram. Build on Cloudflare Workers AI with Deepgram: developers.cloudflare.com/ai-gateway/usa…

Flux Multilingual is live. Real-time conversational speech-to-text for voice agents in 10 languages, with monolingual-grade accuracy, turn detection, and code-switching. Deploy once and launch globally. Learn more → deepgram.com/learn/introduc…