5/5
The sound stack: phone line (@twilio), ears (streaming STT, e.g. @DeepgramAI ), brain ( @cresta Conductor), voice ( @elevenlabs.io ). WebRTC via @LiveKit can save ~300ms.
Past ~300ms, silence feels unnatural. So: when you pause mid-sentence, how does a machine know you're thinking, not done?
The sound stack: phone line (@twilio), ears (streaming STT, e.g. @DeepgramAI ), brain ( @cresta Conductor), voice ( @elevenlabs.io ). WebRTC via @LiveKit can save ~300ms.
Past ~300ms, silence feels unnatural. So: when you pause mid-sentence, how does a machine know you're thinking, not done?
October 9, 2026 at 3:21 AM
5/5
The sound stack: phone line (@twilio), ears (streaming STT, e.g. @DeepgramAI ), brain ( @cresta Conductor), voice ( @elevenlabs.io ). WebRTC via @LiveKit can save ~300ms.
Past ~300ms, silence feels unnatural. So: when you pause mid-sentence, how does a machine know you're thinking, not done?
The sound stack: phone line (@twilio), ears (streaming STT, e.g. @DeepgramAI ), brain ( @cresta Conductor), voice ( @elevenlabs.io ). WebRTC via @LiveKit can save ~300ms.
Past ~300ms, silence feels unnatural. So: when you pause mid-sentence, how does a machine know you're thinking, not done?
Another cool using @TypeSafeAI Jev and @DeepgramAI
Deepgram for turn detection and TTS. Jev picks which prewritten line to say. Every possible response lives in script.ts, written by a human.
The agent cannot say something not approved.
github.com/dg-coreylweathers/jev-voice-agent
Deepgram for turn detection and TTS. Jev picks which prewritten line to say. Every possible response lives in script.ts, written by a human.
The agent cannot say something not approved.
github.com/dg-coreylweathers/jev-voice-agent
GitHub - dg-coreylweathers/jev-voice-agent: A realtime voice agent with no LLM in the loop: Deepgram Flux ends the turn, TypeSafe's Jev picks the reply, Deepgram Flux TTS speaks it.
A realtime voice agent with no LLM in the loop: Deepgram Flux ends the turn, TypeSafe's Jev picks the reply, Deepgram Flux TTS speaks it. - dg-coreylweathers/jev-voice-agent
github.com
September 20, 2026 at 1:50 PM
Another cool using @TypeSafeAI Jev and @DeepgramAI
Deepgram for turn detection and TTS. Jev picks which prewritten line to say. Every possible response lives in script.ts, written by a human.
The agent cannot say something not approved.
github.com/dg-coreylweathers/jev-voice-agent
Deepgram for turn detection and TTS. Jev picks which prewritten line to say. Every possible response lives in script.ts, written by a human.
The agent cannot say something not approved.
github.com/dg-coreylweathers/jev-voice-agent
A voice-first "mini-assistant" that listens, thinks, and speaks back in a DJ R3X-inspired voice from Star Wars.
With @typesafeai Jev & @DeepgramAI Nova-3 streaming plus speculative classification on interim transcripts.
github.com/makeorbreakshop/djr3x_voice
With @typesafeai Jev & @DeepgramAI Nova-3 streaming plus speculative classification on interim transcripts.
github.com/makeorbreakshop/djr3x_voice
GitHub - makeorbreakshop/djr3x_voice
Contribute to makeorbreakshop/djr3x_voice development by creating an account on GitHub.
github.com
September 19, 2026 at 9:59 PM
A voice-first "mini-assistant" that listens, thinks, and speaks back in a DJ R3X-inspired voice from Star Wars.
With @typesafeai Jev & @DeepgramAI Nova-3 streaming plus speculative classification on interim transcripts.
github.com/makeorbreakshop/djr3x_voice
With @typesafeai Jev & @DeepgramAI Nova-3 streaming plus speculative classification on interim transcripts.
github.com/makeorbreakshop/djr3x_voice
As always, thanks to our sponsors 🙏
@meetgranola — The AI notepad for meetings
@DeepgramAI — Enterprise-grade voice AI
@diffusion_io — AI transformation specialists
@AnthropicAI — Claude Code created these video clips
@mercury — The fintech ambitious companies trust
@meetgranola — The AI notepad for meetings
@DeepgramAI — Enterprise-grade voice AI
@diffusion_io — AI transformation specialists
@AnthropicAI — Claude Code created these video clips
@mercury — The fintech ambitious companies trust
September 4, 2026 at 3:28 PM
As always, thanks to our sponsors 🙏
@meetgranola — The AI notepad for meetings
@DeepgramAI — Enterprise-grade voice AI
@diffusion_io — AI transformation specialists
@AnthropicAI — Claude Code created these video clips
@mercury — The fintech ambitious companies trust
@meetgranola — The AI notepad for meetings
@DeepgramAI — Enterprise-grade voice AI
@diffusion_io — AI transformation specialists
@AnthropicAI — Claude Code created these video clips
@mercury — The fintech ambitious companies trust
As always, thanks to our sponsors! 🙏
@mercury – banking built for founders
@claudeai – the AI that created these clips!
@meetgranola – the AI notepad for meetings
@DeepgramAI – lifelike AI voice agents
@diffusion_io – learn to build custom AI software factories
@mercury – banking built for founders
@claudeai – the AI that created these clips!
@meetgranola – the AI notepad for meetings
@DeepgramAI – lifelike AI voice agents
@diffusion_io – learn to build custom AI software factories
August 31, 2026 at 2:47 PM
As always, thanks to our sponsors! 🙏
@mercury – banking built for founders
@claudeai – the AI that created these clips!
@meetgranola – the AI notepad for meetings
@DeepgramAI – lifelike AI voice agents
@diffusion_io – learn to build custom AI software factories
@mercury – banking built for founders
@claudeai – the AI that created these clips!
@meetgranola – the AI notepad for meetings
@DeepgramAI – lifelike AI voice agents
@diffusion_io – learn to build custom AI software factories
Voice AI calls just got a privacy upgrade. 🔒
@DeepgramAI now runs fully on-prem with encrypted audio and model weights, powered by @fortanix Confidential AI and NVIDIA Confidential Computing.
Read the press release: https://nvda.ws/43KdoyN
QT: https://twitter.com/i/status/2061523264439034190
@DeepgramAI now runs fully on-prem with encrypted audio and model weights, powered by @fortanix Confidential AI and NVIDIA Confidential Computing.
Read the press release: https://nvda.ws/43KdoyN
QT: https://twitter.com/i/status/2061523264439034190
June 12, 2026 at 2:05 PM
Voice AI calls just got a privacy upgrade. 🔒
@DeepgramAI now runs fully on-prem with encrypted audio and model weights, powered by @fortanix Confidential AI and NVIDIA Confidential Computing.
Read the press release: https://nvda.ws/43KdoyN
QT: https://twitter.com/i/status/2061523264439034190
@DeepgramAI now runs fully on-prem with encrypted audio and model weights, powered by @fortanix Confidential AI and NVIDIA Confidential Computing.
Read the press release: https://nvda.ws/43KdoyN
QT: https://twitter.com/i/status/2061523264439034190
Can you tell I use reddit almost exclusively for work~
April 16, 2026 at 12:38 AM
Can you tell I use reddit almost exclusively for work~
made a lot of QOL progress with Course Builder today
Uploading videos now provides decently granular feedback as they process in @mux and @DeepgramAI thanks to @inngest and @partykit_io
https://github.com/badass-courses/course-builder/pull/65
Uploading videos now provides decently granular feedback as they process in @mux and @DeepgramAI thanks to @inngest and @partykit_io
https://github.com/badass-courses/course-builder/pull/65
November 4, 2024 at 6:00 PM
made a lot of QOL progress with Course Builder today
Uploading videos now provides decently granular feedback as they process in @mux and @DeepgramAI thanks to @inngest and @partykit_io
https://github.com/badass-courses/course-builder/pull/65
Uploading videos now provides decently granular feedback as they process in @mux and @DeepgramAI thanks to @inngest and @partykit_io
https://github.com/badass-courses/course-builder/pull/65
How @badass_courses built a video publishing pipeline for @kentcdodds's Epic Web
@vercel and @nextjs
@sanity_io content as data
@MuxHQ video
@Cloudflare workers
@DeepgramAI transcripts+subtitles
@OpenAI AI sauce
@inngest event driven workflows✨
https://www.inngest.com/blog/nextjs-trpc-inngest
@vercel and @nextjs
@sanity_io content as data
@MuxHQ video
@Cloudflare workers
@DeepgramAI transcripts+subtitles
@OpenAI AI sauce
@inngest event driven workflows✨
https://www.inngest.com/blog/nextjs-trpc-inngest
November 4, 2024 at 5:14 PM
How @badass_courses built a video publishing pipeline for @kentcdodds's Epic Web
@vercel and @nextjs
@sanity_io content as data
@MuxHQ video
@Cloudflare workers
@DeepgramAI transcripts+subtitles
@OpenAI AI sauce
@inngest event driven workflows✨
https://www.inngest.com/blog/nextjs-trpc-inngest
@vercel and @nextjs
@sanity_io content as data
@MuxHQ video
@Cloudflare workers
@DeepgramAI transcripts+subtitles
@OpenAI AI sauce
@inngest event driven workflows✨
https://www.inngest.com/blog/nextjs-trpc-inngest
working on an article about how we built the video publishing pipeline for @kentcdodds's https://www.epicweb.dev
@cloudinary media
@sanity_io content as data
@MuxHQ for cdn
@Cloudflare workers
@DeepgramAI transcripts & subtitles
@OpenAI AI sauce
✨ @inngest gluing it together
@cloudinary media
@sanity_io content as data
@MuxHQ for cdn
@Cloudflare workers
@DeepgramAI transcripts & subtitles
@OpenAI AI sauce
✨ @inngest gluing it together
November 4, 2024 at 5:12 PM
working on an article about how we built the video publishing pipeline for @kentcdodds's https://www.epicweb.dev
@cloudinary media
@sanity_io content as data
@MuxHQ for cdn
@Cloudflare workers
@DeepgramAI transcripts & subtitles
@OpenAI AI sauce
✨ @inngest gluing it together
@cloudinary media
@sanity_io content as data
@MuxHQ for cdn
@Cloudflare workers
@DeepgramAI transcripts & subtitles
@OpenAI AI sauce
✨ @inngest gluing it together
Révolutionnez la santé avec une IA performante !
Nova-3 Medical, l'API STT dédiée au secteur de la santé.
✅ Précision inégalée : WER de 3.45%
🏥 Adaptée aux environnements cliniques
💰 Économique : À partir de 0.0077$/min
#Santé#IAA#DeepgramAII
March 25, 2025 at 12:00 PM
Révolutionnez la santé avec une IA performante !
Nova-3 Medical, l'API STT dédiée au secteur de la santé.
✅ Précision inégalée : WER de 3.45%
🏥 Adaptée aux environnements cliniques
💰 Économique : À partir de 0.0077$/min
#Santé#IAA#DeepgramAII
Inspired by the incredible @aisdk by @vercel, Orate comes with built-in support for all major TTS / STT providers:
- @OpenAI
- @elevenlabsio
- @AssemblyAI
- @googlecloud
- @IBMwatsonx
- @Azure
- @gladia_io
- @rev
- @MurfAIStudio
- @DeepgramAI
- @SpeechifyAI
- @OpenAI
- @elevenlabsio
- @AssemblyAI
- @googlecloud
- @IBMwatsonx
- @Azure
- @gladia_io
- @rev
- @MurfAIStudio
- @DeepgramAI
- @SpeechifyAI
January 27, 2025 at 4:00 PM
Inspired by the incredible @aisdk by @vercel, Orate comes with built-in support for all major TTS / STT providers:
- @OpenAI
- @elevenlabsio
- @AssemblyAI
- @googlecloud
- @IBMwatsonx
- @Azure
- @gladia_io
- @rev
- @MurfAIStudio
- @DeepgramAI
- @SpeechifyAI
- @OpenAI
- @elevenlabsio
- @AssemblyAI
- @googlecloud
- @IBMwatsonx
- @Azure
- @gladia_io
- @rev
- @MurfAIStudio
- @DeepgramAI
- @SpeechifyAI
Powered by: @lemonslice (live avatars), @elevenlabs @ElevenLabsDevs (voice AI), @livekit (webRTC infra), @deepgramai (voice AI), @ployai (animations), Hamming AI (testing), and many other great companies.
July 28, 2026 at 2:21 AM
Powered by: @lemonslice (live avatars), @elevenlabs @ElevenLabsDevs (voice AI), @livekit (webRTC infra), @deepgramai (voice AI), @ployai (animations), Hamming AI (testing), and many other great companies.
From Y Combinator to working with companies like @DeepgramAI, @perplexity_ai , and @Zoom, it's been a wild ride. We watched voice AI go from an emerging modality to every Fortune 500 enterprise building a conversational interface. Autonomous agents are the future.
June 24, 2026 at 10:24 PM
From Y Combinator to working with companies like @DeepgramAI, @perplexity_ai , and @Zoom, it's been a wild ride. We watched voice AI go from an emerging modality to every Fortune 500 enterprise building a conversational interface. Autonomous agents are the future.
coderlegion.com/5492/one-sma... #abbyconnect #DeepgramAI #AI #SoftwareDevelopment #TechTrends #Coding #AIImplementation #DevLife #TechLeadership #ArtificialIntelligence #ProgrammingTips #TechInnovation
One small company built successful AI with five software engineers—here's what big companies miss.
# Why Most AI Projects Fail (And What Developers Can Learn From One That Didn't) McKinsey says 95% of companies see no ROI from AI. Meanwhile, [Nathan Strum's][1] team at [Abby Connect][2] grew from 3...
coderlegion.com
September 23, 2025 at 11:32 AM
www.insightsfromanalytics.com/post/the-ai-... #abbyconnect #DeepgramAI #AITransformation #BusinessStrategy #FutureOfWork #AISuccess #WorkforceEvolution #TechBusiness #AIImplementation #DigitalTransformation #BusinessInnovation #Leadership
The AI ROI Crisis: Why 95% of Companies Fail (And How One CEO Cracked the Code)
One Las Vegas CEO grew from 30 to 100+ employees while implementing AI. His people-first approach reveals why most companies struggle to achieve AI ROI.McKinsey reports that 95% of companies see littl...
www.insightsfromanalytics.com
September 22, 2025 at 8:12 PM
🧠 Just added Deepgram to EveryDev!
Speech-to-text, text-to-speech, and a full voice agent API — all from the folks at @deepgramai.
Perfect for building voice UIs, real-time assistants, or audio analytics.
🎙️ speech recognition
🗣️ voice synthesis
🤖 voice agent API
www.everydev.ai/tools/deepgram
Speech-to-text, text-to-speech, and a full voice agent API — all from the folks at @deepgramai.
Perfect for building voice UIs, real-time assistants, or audio analytics.
🎙️ speech recognition
🗣️ voice synthesis
🤖 voice agent API
www.everydev.ai/tools/deepgram
Deepgram | EveryDev.ai
Deepgram provides a suite of developer APIs for voice and audio processing, including Speech-to-Text, Text-to-Speech, Voice Agent API, and Audio…
www.everydev.ai
August 2, 2025 at 1:02 AM
🧠 Just added Deepgram to EveryDev!
Speech-to-text, text-to-speech, and a full voice agent API — all from the folks at @deepgramai.
Perfect for building voice UIs, real-time assistants, or audio analytics.
🎙️ speech recognition
🗣️ voice synthesis
🤖 voice agent API
www.everydev.ai/tools/deepgram
Speech-to-text, text-to-speech, and a full voice agent API — all from the folks at @deepgramai.
Perfect for building voice UIs, real-time assistants, or audio analytics.
🎙️ speech recognition
🗣️ voice synthesis
🤖 voice agent API
www.everydev.ai/tools/deepgram
@richiemcilroy https://x.com/richiemcilroy/status/1938659588422709462 #x-richiemcilroy
hey @DeepgramAI, can you guys please take a look at my DM. ty 🙏
hey @DeepgramAI, can you guys please take a look at my DM. ty 🙏
June 27, 2025 at 6:30 PM
@richiemcilroy https://x.com/richiemcilroy/status/1938659588422709462 #x-richiemcilroy
hey @DeepgramAI, can you guys please take a look at my DM. ty 🙏
hey @DeepgramAI, can you guys please take a look at my DM. ty 🙏
@rauchg https://x.com/rauchg/status/1865125266336432271 #x-rauchg 🎥
We're building a passive AI fact checker with @aisdk
A live fake news detector if you will.
Good demo of realtime ai. Powered by @perplexity_ai & @deepgramai – open sourced soon!
We're building a passive AI fact checker with @aisdk
A live fake news detector if you will.
Good demo of realtime ai. Powered by @perplexity_ai & @deepgramai – open sourced soon!
December 6, 2024 at 8:30 PM
@rauchg https://x.com/rauchg/status/1865125266336432271 #x-rauchg 🎥
We're building a passive AI fact checker with @aisdk
A live fake news detector if you will.
Good demo of realtime ai. Powered by @perplexity_ai & @deepgramai – open sourced soon!
We're building a passive AI fact checker with @aisdk
A live fake news detector if you will.
Good demo of realtime ai. Powered by @perplexity_ai & @deepgramai – open sourced soon!
Deepgram launched Nova-3 Medical, a specialized speech-to-text model for healthcare
x.com/DeepgramAI/...
x.com/DeepgramAI/...
March 4, 2025 at 10:35 AM
Deepgram launched Nova-3 Medical, a specialized speech-to-text model for healthcare
x.com/DeepgramAI/...
x.com/DeepgramAI/...
Last time I built a Voice AI project with @DeepgramAI.
What voice AI platform or framework should I try next?
What voice AI platform or framework should I try next?
July 4, 2026 at 7:45 PM
Last time I built a Voice AI project with @DeepgramAI.
What voice AI platform or framework should I try next?
What voice AI platform or framework should I try next?
Build local coding agents with Cursor harness.
VoxCode is a voice-powered coding workspace that runs entirely on your machine
✅ Features:
→ Voice codebase summaries
→ Architecture Q&A
→ Voice-based file edits
→ Live agent activity logs in the UI
Built with @DeepgramAI, @nebiustf, and @cursor_ai
VoxCode is a voice-powered coding workspace that runs entirely on your machine
✅ Features:
→ Voice codebase summaries
→ Architecture Q&A
→ Voice-based file edits
→ Live agent activity logs in the UI
Built with @DeepgramAI, @nebiustf, and @cursor_ai
June 11, 2026 at 5:07 PM
Build local coding agents with Cursor harness.
VoxCode is a voice-powered coding workspace that runs entirely on your machine
✅ Features:
→ Voice codebase summaries
→ Architecture Q&A
→ Voice-based file edits
→ Live agent activity logs in the UI
Built with @DeepgramAI, @nebiustf, and @cursor_ai
VoxCode is a voice-powered coding workspace that runs entirely on your machine
✅ Features:
→ Voice codebase summaries
→ Architecture Q&A
→ Voice-based file edits
→ Live agent activity logs in the UI
Built with @DeepgramAI, @nebiustf, and @cursor_ai
🚀 Exciting news! @DeepgramAI & @IBM are teaming up to redefine enterprise voice AI! With real-time speech capabilities, businesses can now unlock #AI-driven customer support like never before!
Deepgram and IBM Introduce Advanced Voice Capabilities for Enterprise AI
Promoting the collaboration between Deepgram and IBM, which enhances enterprise AI with advanced voice capabilities for real-time applications.
deepgram.com
February 24, 2026 at 1:38 PM
🚀 Exciting news! @DeepgramAI & @IBM are teaming up to redefine enterprise voice AI! With real-time speech capabilities, businesses can now unlock #AI-driven customer support like never before!
🚀 Introducing Nova-3! The most advanced voice AI, delivering unmatched accuracy even in noisy environments! Join industry leaders using it for real-time multilingual support! 📈 #DeepgramAI #VoiceAI
Introducing Nova-3: Extending Deepgram's Leadership in Voice AI for Enterprise Use Cases
Deepgram announces Nova-3, its latest speech-to-text model, designed for enterprise use with superior accuracy in challenging audio conditions, real-time multilingual support, and self-service customization.
deepgram.com
February 14, 2025 at 8:03 PM
🚀 Introducing Nova-3! The most advanced voice AI, delivering unmatched accuracy even in noisy environments! Join industry leaders using it for real-time multilingual support! 📈 #DeepgramAI #VoiceAI