NVIDIA

Technology

Daily is a WebRTC infrastructure provider for video and audio calls, and Deepgram is the recommended speech provider inside the Daily ecosystem. Live transcription, meeting summaries, voice agent pipelines, and live captions all run on Deepgram models behind a few lines of Daily SDK configuration. For product teams, this matters because adding real-time voice intelligence to a video product (telehealth, customer support, recruiting, training) usually means evaluating a transcription vendor, building an audio routing layer, and writing reconciliation logic. Daily and Deepgram remove most of that. The audio is already streaming through Daily; pointing it at Deepgram is one configuration step: pass enable_transcription with the value "Deepgram:<your API key>" and live transcription is enabled at the account level. Real-time streaming uses WebSockets so new audio streams to Deepgram while the latest transcription results stream back, full duplex. The pattern supports live captions for accessibility, real-time meeting transcription with diarization, and any use case where latency matters. For voice agents specifically, Pipecat (Daily's open-source voice agent framework) ships with native Deepgram services. It's the production path from a working prototype to a deployed agent. If you are building on Daily and want voice intelligence, the integration walks through the SDK property in a few minutes. For voice agents, see the Pipecat partner page. Outlinks & Resources Daily.co partner page Add Live Transcriptions to a Daily Video Call with Deepgram (Built with Deepgram) Video: Add Live Transcriptions to a Daily Video Call With Deepgram Transcribe Meetings in Realtime (developer docs) Pipecat (Daily's voice agent framework, GitHub) Daily WebRTC API Deepgram documentation Contact Deepgram

Technology

Vapi is one of the leading voice agent platforms, and Deepgram provides the speech layer inside it. Builders who pick Vapi to ship a production voice agent get Deepgram's Nova-3 STT, Flux for conversational turn detection, and Aura-2 TTS as native model options, configured in a few clicks rather than a custom integration. Vapi handles 62M+ calls monthly with 99.99% uptime, making it the proven choice for enterprises requiring scale across voice, SMS, and chat. Deepgram is integrated for its speed and human-like voice quality, and powers the STT and TTS surface that real production voice agents depend on. Voice agent quality lives in the speech layer. The transcription needs to handle interruptions, background noise, accents, and the alphanumeric read-back that real conversations include. The TTS needs to sound natural at low enough latency that the caller does not notice. Deepgram's voice-native foundational models are built specifically for those requirements, and Vapi exposes them as the platform's recommended models for production agents. The pattern: a developer picks Vapi to handle orchestration (transport, interruption, tool calls, state), picks Deepgram models for STT and TTS, drops in an LLM, and ships a working voice agent in hours. For latency-critical or regulated workloads, the same Vapi configuration runs against Deepgram's self-hosted runtime, keeping audio inside the customer's environment. If you are building a voice agent on Vapi and want production-grade speech, Deepgram is the default. Resources below. Outlinks & Resources Vapi partner page Vapi: Voice AI Platform for Developers (Voice AI Apps) Top Voice AI Agents for 2026: The Ultimate Buyer's Guide Vapi platform Deepgram Voice Agent API Contact Deepgram

Technology

Communications / CPaaS

Speech to Text

Text to Speech

STT Nova

TTS Flux

Deepgram and LiveKit enable developers to build production-ready voice AI agents with real-time speech recognition and natural text-to-speech, all within LiveKit's open-source framework for voice, video, and physical AI agents. LiveKit's Agents framework ships native Deepgram plugins for both STT and TTS. Developers run Nova-3 for high-accuracy streaming transcription across 30+ languages, or Flux for conversational speech recognition with built-in end-of-turn detection and interruption handling. Flux is purpose-built for voice agent workflows where handling barge-in cleanly is the difference between a demo and a production deployment. Model selection is a single configuration line in Python or TypeScript. The integration covers the core voice agent use cases: customer support automation, multilingual voice assistants, meeting bots, and inbound/outbound telephony via LiveKit's SIP integration. Deepgram models are also available through LiveKit Inference, with billing and model access managed directly through LiveKit's platform. LiveKit runs on a global cloud network built for production workloads, with over 300,000 developers and billions of calls processed annually. With voice-native infrastructure and Deepgram's sub-300ms latency speech models, development teams get the fastest path from prototype to production for agents that hold real conversations at scale. Outlinks & Resources LiveKit Agents Framework (Docs): https://docs.livekit.io/agents/ LiveKit Agents Framework (GitHub): https://github.com/livekit/agents LiveKit Deepgram STT Plugin Guide: https://docs.livekit.io/agents/models/stt/deepgram/ LiveKit Deepgram TTS Plugin Guide: https://docs.livekit.io/agents/integrations/tts/deepgram/ LiveKit Deepgram Integration Guide: https://developers.deepgram.com/docs/livekit-integration LiveKit Integration Tutorial: https://developers.deepgram.com/docs/build-voice-agent-with-livekit-and-deepgram Build a Voice Agent with LiveKit and Deepgram: https://developers.deepgram.com/docs/build-voice-agent-with-livekit-and-deepgram

Technology

IBM selected Deepgram as its first fully embedded third-party speech model provider after customers asked for higher-accuracy STT and TTS, lower latency, and more flexible deployment than traditional cloud speech services could deliver. Deepgram's speech-to-text and text-to-speech models are now embedded into IBM watsonx, giving enterprises a modern voice AI stack on top of IBM's infrastructure. By embedding Deepgram inside watsonx Orchestrate Agent Builder, IBM clients can build voice agents and voice-enabled workflows on top of a real-time speech foundation that has been refined over more than a decade. Customers tap Deepgram's STT and TTS capabilities directly through their existing IBM environment, without adding extra vendors or commercial complexity. The integration covers contact-center conditions (accents, background noise, alphanumeric read-back) where speech accuracy makes the call usable, latency low enough to feel like a real conversation, and deployment across IBM Cloud, customer VPC, and self-hosted on regulated infrastructure. The integration is announced via joint press release and shipping inside watsonx Orchestrate today. If you are running IBM watsonx Orchestrate and evaluating voice, the integration is built in. Reach out via deepgram.com/contact-us for enterprise terms. Outlinks & Resources Deepgram and IBM Introduce Advanced Voice Capabilities for Enterprise AI Deepgram × IBM: Enterprise Voice AI Inside watsonx CX IBM watsonx Orchestrate Powered by Deepgram Contact Deepgram

Technology

Real-time call transcription Twilio Media Streams pipes real-time call audio to Deepgram for transcription. The reference architecture forks audio from calls to a proxy server via WebSockets, the server forwards to Deepgram, and transcripts flow back to subscribed clients. Setup needs a Twilio number, a Deepgram API key, and ngrok for local development. Transcribing Twilio Voice Calls in Real-Time with Deepgram Transcribe Twilio Voice Calls in Real-Time with Rust and Deepgram Twilio and Deepgram STT (developer docs) On-premise Twilio integration tools Recorded call transcription For batch transcription of completed calls, Programmable Voice recordings flow into Deepgram via API. Transcribe Recorded Calls With Twilio (developer docs) Text-to-speech for the response side Aura-2 handles speech synthesis for AI-generated responses on the outbound side of a Twilio call. Twilio and Deepgram TTS (developer docs) Voice agents on Twilio Full voice agent flows orchestrate STT, an LLM, and TTS through Twilio. Documented patterns include OpenAI as the LLM and the Deepgram Voice Agent API as the speech layer. Build a Voice Agent with Twilio, OpenAI, and Deepgram (developer docs) Twilio and Deepgram Voice Agent (developer docs) Workflow automation (Twilio Studio + Pipedream) For non-developer ops teams, Twilio Studio plus Pipedream plus Deepgram delivers automated transcription pipelines with email or webhook delivery. Email Phone Call Transcripts with Twilio Studio and Pipedream Get started If you are building voice features on Twilio, the fastest start is the Marketplace Add-On. For the integration patterns above, the developer docs walk through each end to end. Enterprise terms via deepgram.com/contact-us. Outlinks & Resources Twilio Marketplace Add-On for Deepgram (find Deepgram listing) Twilio Media Streams documentation Deepgram documentation Contact Deepgram

Technology

Enterprise Voice AI, Native to AWS. Real-time speech-to-text, text-to-speech, and voice agents that run inside your AWS stack: Amazon Connect, SageMaker, EC2, Bedrock and more. For background on the partnership at scale, see AWS and Deepgram: The Foundation That Makes Voice AI Scale. Procurement Deepgram purchases draw down on your existing AWS commit. Nova-3 STT, Aura-2 TTS, and the Voice Agent API are available through AWS Marketplace and AWS credits available for enterprise procurement. Product integrations Amazon Connect. Deepgram is a productized speech provider inside Amazon Connect contact flows. For contact center teams running Connect, Nova-3 replaces or augments Connect's native STT for live transcription, and voicebot use cases, with a 30% average WER improvement over alternative STT options. The integration also extends to Amazon Lex for IVR and virtual voice agents. Amazon SageMaker. Deepgram STT and TTS are productized for SageMaker via dedicated Model Packages on AWS Marketplace, so customers can deploy real-time speech inference inside their own VPC. See Deployment options below for the SageMaker deployment path. Amazon Bedrock. Deepgram's Voice Agent API integrates natively with Amazon Bedrock as the LLM layer, giving enterprises a single procurement and inference surface for both Deepgram speech models and the Bedrock-hosted LLMs (Claude, Titan, AI21) they pair with. Two integration paths are supported: native via the built-in aws_bedrock provider type, and a proxy-server option for advanced cases. Deployment options Managed API. Deepgram's hosted API consumed directly from any AWS workload. The fastest path for teams that want to start shipping today. PrivateLink and VPC Endpoints. Managed Deepgram with no public-internet exposure. Audio routes from the customer's VPC to Deepgram's service endpoints over the AWS private backbone, suitable for security-conscious enterprises and regulated workloads. Amazon SageMaker. Deepgram STT and TTS models deployed inside the customer's own VPC using SageMaker Model Packages. Subscribe to Deepgram in AWS Marketplace, stand up a SageMaker Endpoint, and tune environment variables for concurrency, interim results, and streaming behavior. The middle ground between fully-managed Deepgram and full self-hosted operations: the data plane stays in the customer's account, but SageMaker handles the deployment and scaling surface. Self-hosted on AWS. For customers who need full control of the runtime, Deepgram's self-hosted models deploy on EC2, EKS, ECS, and across the G5, G6, and G6e instance families with production-grade operations. Each GPU instance can run many concurrent real-time streams. For regulated industries, this means a single AWS account becomes the home for the entire voice stack: contact center, healthcare workflows, media transcription, and voice agents. Get started If you are evaluating voice AI on AWS, the fastest path is the Marketplace listing. For enterprise terms, ISV Accelerate co-sell, or specific Connect, SageMaker, or Bedrock integrations, deepgram.com/contact-us is the direct route. Outlinks & Resources Procurement and partnership Deepgram Achieves AWS Generative AI Competency AWS and Deepgram: The Foundation That Makes Voice AI Scale How Enterprises Can Unlock Voice AI Through the Deepgram Partner Ecosystem AWS Marketplace listing Amazon Connect + Lex Deepgram and AWS Amazon Connect Integration to Unlock Voice Data at Scale Deepgram Brings Best-in-Class Voice AI Models to Amazon Connect & Lex Amazon Connect and Deepgram (developer docs) Amazon SageMaker Deepgram Brings Real-Time Speech Intelligence to Amazon SageMaker Deploy Deepgram on Amazon SageMaker (developer docs) Configure Amazon SageMaker Deployments (developer docs) Webinar: How to Deploy Real-Time Voice AI Within SageMaker Amazon Bedrock Announcing Deepgram Voice Agent API Integration with AWS Bedrock Amazon Bedrock and Deepgram Voice Agent (developer docs) AWS Lambda + serverless Deploy a Serverless Transcription Workflow with AWS Lambda + Deepgram STT AWS S3 Presigned URLs and Deepgram (developer docs) Self-hosted on AWS Self-Hosted on AWS using Docker or Podman (developer docs) Deployment Environments (developer docs) Demos, case studies, comparison Deepgram + AWS Voice AI Demo: Real-Time Medical Speech-to-Text in Action A Year of Voice AI at Scale Migrating From AWS Transcribe to Deepgram (developer docs) Compare Amazon Transcribe API Alternatives Contact Deepgram documentation Contact Deepgram

Technology

Healthcare

Speech to Text

STT Nova

Deepgram is bringing enterprise-grade speech recognition directly onto PCs powered by Snapdragon® processors. By optimizing Deepgram's Nova-3 speech-to-text model on the Qualcomm® Hexagon™ NPU in the Snapdragon X Series platform, Deepgram is enabling developers and device manufacturers to deliver real-time voice experiences with greater speed, privacy, and reliability, without relying on a cloud connection. Many voice AI solutions have historically relied on a cloud-based architecture. Before a response could be delivered, each interaction required audio to leave the device, travel to the cloud, be processed somewhere else, and then return. With this approach, delays and privacy concerns are sometimes introduced, which limit where voice AI can realistically be deployed. Deepgram is fundamentally changing that model, enabling speech recognition to happen directly on the device itself. The result is an entirely new world of applications and user experiences that feel like a natural conversation, whether it is running in a vehicle, on an AI PC, inside an XR headset, or at the edge of a network where connectivity cannot be guaranteed. Nova-3 advances Deepgram's industry-leading accuracy, extending its capabilities to a broader range of real-world enterprise use cases and challenging audio conditions. It is the first voice AI model to offer real-time multilingual transcription, and the first to provide demonstrably effective self-serve customization, enabling instant vocabulary adaptation without model retraining. In Deepgram's benchmark, Nova-3 delivers the lowest word error rate of any model tested. This effort opens the door to further integrating voice into a new generation of intelligent applications across automotive, mobile, AI PC, XR, industrial edge, IoT, and wearable devices. If you are building for PCs powered by Snapdragon and evaluating on-device speech recognition, deepgram.com/contact-us is the direct path. Outlinks & Resources Deepgram Delivers Real-Time Voice AI at the Edge for use with Snapdragon (press release) Snapdragon PCs & Laptops Product List | Snapdragon Deepgram Nova-3 Speech-to-Text Contact Deepgram Qualcomm, Hexagon and Snapdragon are trademarks or registered trademarks of Qualcomm Incorporated. Snapdragon and Qualcomm branded products are products of Qualcomm Technologies, Inc. and/or its subsidiaries.

Carahsoft is the procurement bridge between Deepgram and the public sector. Through Carahsoft's contract vehicles, federal agencies, military commands, state and local governments, education customers, and healthcare buyers can acquire Deepgram on terms that work with their procurement process. Public sector use cases that matter: federal agency call-center transcription for compliance and FOIA responsiveness; DoD contact center voice AI on FedRAMP-authorized infrastructure; veterans services voice agents with multilingual STT; state and local 311 transcription for analytics; education voice automation for student services; healthcare voice AI for HIPAA-aligned PHI handling. The deployment posture matches the procurement posture. Deepgram supports SOC 2, HIPAA, and standard public-sector compliance frameworks. The self-hosted runtime runs in customer-controlled environments (agency cloud, AWS GovCloud, on-premises, and air-gapped) with no audio leaving the protected boundary. For workloads requiring confidential computing (sovereign cloud, classified, or CUI handling), Deepgram pairs with Fortanix for hardware-enforced isolation. If you are a public sector buyer evaluating voice AI, Carahsoft handles the procurement path and Deepgram provides the speech layer. Carahsoft is The Trusted IT Solutions Provider®, supporting Federal, State, Local Government, Education, Healthcare, and Private Sector industries. As the Master Aggregator™ for our vendor, reseller and integrator partners, we deliver solutions for Cybersecurity, FinOps, MultiCloud, DevSecOps, Artificial Intelligence, Human Capital, Legal & Courtroom Technology, Customer Experience & Engagement, and more. Carahsoft is consistently recognized by its partners as a top revenue producer and is listed annually among the industry's fastest growing firms by CRN, Inc., Forbes, Inc. 5000, and Washington Business Journal. Founded in 2004, Carahsoft is headquartered in Reston, VA. Visit our site. Outlinks & Resources Carahsoft Deepgram page Deepgram Self-Hosted overview Self-Hosted developer docs Speech-to-Text Privacy: Enterprise Security, Compliance and Data Protection Standard Compliance: HIPAA, SOC 2, GDPR Information Security & Privacy Statement Data Privacy Compliance Contact Deepgram

Technology

Deepgram brings its Voice AI Platform to the Dell AI Factory with NVIDIA, giving enterprises a path to run speech-to-text, text-to-speech, and voice agents on Dell infrastructure inside their own data centers. The stack pairs Deepgram's models with NVIDIA accelerated computing on Dell servers, operated through the Dell AI Factory model. For security-sensitive workloads, Fortanix Confidential AI adds model-in-use protection on NVIDIA Confidential Computing-enabled GPUs, keeping audio data and model weights encrypted throughout processing. This gives regulated and data-sovereign organizations a supported route to production voice AI without sending audio to a public cloud. A joint reference architecture is in development so teams can deploy a validated configuration rather than assembling one from scratch. If you are planning an on-premises voice AI deployment on Dell infrastructure, reach out via deepgram.com/contact-us to discuss deployment options. Outlinks & Resources Dell AI Factory Deepgram × Fortanix: Private Voice AI for Regulated Industries Deepgram Delivers Private Voice AI to Regulated Industries (June 2026 press release) Contact Deepgram

Technology

Communications / CPaaS

Cloud Infrastructure

Speech to Text

Global

Deepgram and Fortanix bring private voice AI to regulated industries. The partnership, announced in June 2026, lets enterprises run Deepgram's voice AI models entirely within their own environment while their most sensitive data stays protected. The deployment runs Deepgram's models with Fortanix Confidential AI on NVIDIA Confidential Computing-enabled GPUs. Workloads execute inside a hardware-isolated trusted execution environment, a protection traditional infrastructure cannot provide. Audio data and Deepgram's proprietary model weights remain encrypted throughout active use, which protects customer data on one side and guards the models against theft or inappropriate use on the other. For healthcare systems, financial institutions, and public-sector organizations, this closes the gap between voice AI adoption and security mandates. Deployments can meet HIPAA, GDPR, and national data-residency requirements without compromising on model quality. In Deepgram's benchmark, Nova-3 delivers the lowest word error rate of any model tested, and real-time transcription runs at sub-300ms latency inside the customer's own environment. If you are evaluating voice AI for a regulated or data-sovereign environment, reach out via deepgram.com/contact-us to discuss on-premises deployment options. Outlinks & Resources Deepgram Delivers Private Voice AI to Regulated Industries with On-Premises Deployments Powered by Fortanix Confidential AI and NVIDIA Confidential Computing Fortanix Confidential AI NVIDIA Confidential Computing Contact Deepgram

Development

Contact Center

Healthcare

Speech to Text

Audio Intelligence

Global

JetBridge is a custom software engineering firm founded by the engineers behind Five9 and DoctorBase. The team builds production-grade voice AI applications for VC-backed startups and mid-market companies, with deep experience in contact center and healthcare environments where accuracy and compliance requirements leave no margin for a mediocre speech layer. JetBridge integrates Deepgram's models as core components in the applications it builds for clients. Nova-3 handles real-time transcription for contact center and CX workflows; Nova-3 Medical covers clinical documentation and healthcare applications where domain-specific vocabulary and HIPAA compliance are requirements, not optional. The team has applied this stack for companies backed by Andreessen Horowitz, including Asimov, and for enterprise clients including LabCorp and Compass. JetBridge runs on a Forward Deployed Engineering model: clients see a working proof of concept before committing to a full engagement. That approach compresses the timeline between architecture review and working software, which matters for startups building toward a fundraise and for mid-market companies running a competitive evaluation. There is no upfront cost to validate the approach before the partnership begins. Teams evaluating Deepgram for a custom integration can reach JetBridge directly at deepgram@jetbridge.com for a free architecture consultation, including cost and timeline estimates. JetBridge responds within minutes. Outlinks & Resources JetBridge: https://jetbridge.com Contact JetBridge: deepgram@jetbridge.com Deepgram Enterprise: https://deepgram.com/enterprise Deepgram Documentation: https://developers.deepgram.com

Channel Partner

Contact Center

Healthcare

Financial Services

Professional Services

Text to Speech

Voice Agent

TTS Flux

Global

Cognizant is one of the world's largest IT services and consulting firms, with more than 300,000 professionals delivering AI, cloud, and digital transformation programs across every major industry. The company integrates Deepgram's voice AI models into the enterprise AI solutions it architects and deploys for its clients. Cognizant's AI practice draws on Deepgram for the speech layer in AI-driven applications: real-time contact center transcription, voice-enabled AI agents, compliance call recording, and speech interfaces for enterprise workflows. Nova-3 handles high-accuracy streaming transcription across Cognizant's core verticals. Flux handles conversational STT in voice agent deployments where built-in turn detection and interruption handling are requirements, not add-ons. The industries Cognizant serves are where voice AI carries the most operational weight. Financial services clients need accurate, auditable call transcription feeding QA and risk systems. Healthcare clients need clinical documentation that meets HIPAA requirements and handles domain-specific vocabulary. Contact center clients need AI agents that can sustain real conversations at volume. Cognizant's Neuro AI platform and AI-led modernization practice provide the implementation scale to move voice AI from proof of concept to production across these environments. Enterprises looking to add voice AI capabilities to a broader AI transformation program can engage Cognizant through their standard advisory process or contact Deepgram's enterprise team to explore the partnership. Outlinks & Resources Cognizant AI Services: https://www.cognizant.com/us/en/services/ai Cognizant Neuro AI Platform: https://www.cognizant.com/us/en/services/ai/enterprise-ai Deepgram Enterprise: https://deepgram.com/enterprise Deepgram Documentation: https://developers.deepgram.com Contact Deepgram: https://deepgram.com/contact-us

Accenture is one of the largest global systems integrators, with deep practices across contact center transformation, customer experience, enterprise generative AI, and public sector technology delivery. Deepgram provides the speech-to-text and text-to-speech layer underneath the voice AI workloads Accenture's clients are building. For enterprises evaluating a partner to bring voice AI into production, the practical questions are about workflow design, integration with existing telephony and CCaaS stacks, and day-two operations. Accenture brings the strategy, integration, and managed-services capacity for that work. Deepgram brings the speech models: Nova-3 for streaming STT, Aura-2 for TTS, and a self-hosted runtime for regulated workloads where audio cannot leave the customer's environment. The combined offering covers contact center modernization (live transcription, agent assist, automated quality management, post-call analytics), enterprise voice agents (real-time STT, conversational turn detection, low-latency TTS), and regulated-industry deployments (healthcare, financial services, public sector) where compliance posture and data residency are required from day one. If you are working with Accenture and evaluating Deepgram, or building a voice AI program and want an implementation partner who can ship into production, deepgram.com/contact-us is the direct path. Outlinks & Resources How Enterprises Can Unlock Voice AI Through the Deepgram Partner Ecosystem Deepgram Self-Hosted overview Deepgram documentation Contact Deepgram

500 Global is one of the largest seed-stage VC firms and accelerators, with roughly $2.3B AUM, more than 2,500 portfolio companies across 80+ countries, and 38+ unicorns. Deepgram is now on 500 FounderHub, the perks platform 500 Global runs for its portfolio. For 500 portfolio companies building voice products (AI agents, contact center software, real-time transcription, voice features in consumer or vertical SaaS), Deepgram provides the speech layer underneath. Nova-3 for streaming speech-to-text, Aura-2 for text-to-speech, and the Voice Agent API for full agent workflows are available through the FounderHub benefit. If you are a 500 Global founder, the path to claim Deepgram credits runs through your FounderHub login. For terms beyond the program tier or for enterprise procurement, deepgram.com/contact-us is the direct route. Outlinks & Resources 500 Global 500 FounderHub Deepgram Startup Program Deepgram documentation Contact Deepgram

Technology

8x8's XCaaS (Experience Communications as a Service) brings cloud contact center, voice, video, chat, and APIs together in a single integrated offering, and Deepgram provides the speech-to-text layer inside it. Real-time transcription, call analytics, and voice intelligence inside 8x8 run on Deepgram's Nova-3 model. Speech accuracy has to hold up in the conditions an 8x8 customer actually runs: diverse accents, background noise, and the domain-specific vocabulary (account numbers, policy IDs, product SKUs) that drives self-service interactions. Nova-3 is built for that envelope, with low enough latency to support agent assist on live calls and clean enough transcripts to feed downstream quality-management and analytics workflows. For enterprises running 8x8 Contact Center, the integration covers the surfaces that depend on speech: live transcription for agent assist, automated QM scoring, and post-call conversational analytics. 8x8 handles the call routing, workforce engagement, and CX surface; Deepgram provides speech recognition designed for real-time production use. If you are running 8x8 and evaluating speech recognition, deepgram.com/contact-us is the direct path. Outlinks & Resources 8x8 Platform 8x8 Contact Center Deepgram documentation Contact Deepgram

Technology

Cisco operates some of the largest enterprise voice infrastructure in the world, from contact center platforms to collaboration and unified communications. Deepgram provides streaming speech-to-text for Cisco voice workloads where transcription accuracy and latency matter, with Nova-3 as the underlying model. Real-time speech recognition has to hold up in the conditions Cisco customers actually run: diverse accents, background noise, domain-specific vocabulary, and alphanumeric read-back where a single misheard digit breaks the call. Nova-3 is built for that production envelope, and Deepgram's voice-native foundational models are designed for the throughput a large contact center or voice-enabled product requires. For Cisco IVR platforms, Deepgram integrates through the open-source UniMRCP protocol, which lets enterprises drop Deepgram STT into existing Cisco voice flows without re-architecting the call path. The plugin and developer documentation are published below. If you are a Cisco customer or product team evaluating speech recognition, deepgram.com/contact-us is the direct route. Outlinks & Resources Deepgram Announces UniMRCP Integration to Power Modern Customer Experience Deepgram UniMRCP Plugin (developer docs) Deepgram UniMRCP server plugin (GitHub) Deepgram documentation Contact Deepgram