1. [Topics](https://developer.nvidia.com/topics/) [AI](https://developer.nvidia.com/topics/ai) [Generative AI](https://developer.nvidia.com/topics/ai/generative-ai/) NVIDIA Riva # NVIDIA Nemotron Speech for Developers [NVIDIA Nemotron](/topics/ai/nemotron)™ Speech is a family of GPU-accelerated open models for automatic speech recognition (ASR), text-to-speech (TTS), neural machine translation (NMT), and speech-to-speech across ~40 languages. Deployed through the NVIDIA Riva library for optimized inference, the speech models are fully customizable and run across all clouds, in data centers, at the edge, and in embedded devices. As speech becomes the primary interface for AI agents, these models enable organizations to build voice-first AI agents that understand and respond naturally across dozens of languages. [Try APIs](https://build.nvidia.com/explore/speech "Download Now")[Documentation](https://docs.nvidia.com/deeplearning/riva/user-guide/docs/index.html "Forums link")[Forum](https://forums.developer.nvidia.com/c/ai-data-science/deep-learning/riva/475 "Go to Documentation") * * * ## How Nemotron Speech Works Speech and translation AI microservices convert spoken words into text (speech recognition), written language into spoken words (speech synthesis), and spoken or written words from one language to another (translation). Pretrained AI models are trained on vast datasets and can be fine-tuned on custom datasets to accelerate the development of domain-specific models. Fully containerized, these microservices are optimized for real-time performance and offline high throughput on premises or in the cloud and can quickly scale to hundreds and thousands of parallel streams. ![A workflow diagram showing speech and translation AI microservices](https://developer.download.nvidia.com/images/products/riva-how-it-works-2560x940.svg) ### Quick-Start Guide Get step-by-step instructions for deploying pretrained models and how to interact with them. [Get Started ](https://docs.nvidia.com/deeplearning/riva/user-guide/docs/quick-start-guide.html?_gl=1*mut5vc*_gcl_au*NDk1OTAwNzg1LjE3MTYyMjY5Njc.) ### Learn to Add Voice to Your Apps Learn how to add real-time speech recognition, translation, and natural-sounding text-to-speech to your applications using NVIDIA NIM™ for production-ready voice experiences. [Read Blog](https://developer.nvidia.com/blog/quickly-voice-your-apps-with-nvidia-nim-microservices-for-speech-and-translation/) ### Tutorial Read how to build a voice-powered AI agent that combines the NVIDIA Nemotron streaming ASR model with multimodal retrieval-augmented generation (RAG), safety guardrails, and long-context reasoning. [Read Blog ](https://developer.nvidia.com/blog/how-to-build-a-voice-agent-with-rag-and-safety-guardrails/) ### Video Demo Learn how to build a voice-enabled AI agent that integrates an NVIDIA Nemotron ultra-low-latency streaming model for real-time automatic speech recognition. [Watch Video](https://www.youtube.com/watch?v=bCDcziL9OCo) * * * ## Ways to Get Started With Nemotron Speech Use the right tools and technologies to build and deploy fully customizable, multilingual speech and translation AI applications. ![A decorative image of building AI application with NVIDIA NIM APIs](https://developer.download.nvidia.com/icons/m48-nim-256px-blk%201@2x.png) ### Try Experience Nemotron Speech through a UI-based portal for exploring and prototyping with NVIDIA-managed endpoints, available for free through [build.nvidia.com](http://build.nvidia.com). [Try Now](https://build.nvidia.com/explore/speech) ![A decorative image of deploying with NVIDIA AI Enterprise](https://developer.download.nvidia.com/icons/m48-digital-deep-learning-institute-talks-training.svg) ### Deploy Get a free license to try [NVIDIA AI Enterprise](https://www.nvidia.com/en-us/data-center/products/ai-enterprise/) for 90 days using your existing infrastructure. [Request a 90-Day License](https://enterpriseproductregistration.nvidia.com/?LicType=EVAL&ProductFamily=NVAIEnterprise) * * * ## Development Starter Kits Start developing your speech and translation AI application with Nemotron Speech by accessing tutorials, notebooks, [forums](https://forums.developer.nvidia.com/c/ai-data-science/deep-learning/riva/475), [release notes](https://docs.nvidia.com/deeplearning/riva/user-guide/docs/release-notes.html), and comprehensive [documentation](https://docs.nvidia.com/deeplearning/riva/user-guide/docs/index.html). ### Automatic Speech Recognition Achieve high transcription accuracy for Arabic, English, French, German, Hindi, Italian, Japanese, Korean, Mandarin, Portuguese, Russian, and Spanish with state-of-the-art models pretrained on thousands of hours of audio on NVIDIA supercomputers. - [Read Speech-Recognition Documentation](https://docs.nvidia.com/deeplearning/riva/user-guide/docs/asr/asr-overview.html) - [Read Quick-Start Guide](https://docs.nvidia.com/deeplearning/riva/user-guide/docs/quick-start-guide.html) - [Explore Speech-Recognition Tutorials](https://docs.nvidia.com/deeplearning/riva/user-guide/docs/tutorials-asr-contents.html) ### Text-to-Speech Customize across English, German, Italian, Mandarin, and Spanish TTS pipelines for the voice and intonation you want. - [Read TTS Documentation](https://docs.nvidia.com/deeplearning/riva/user-guide/docs/tts/tts-overview.html) - [Read Quick-Start Guide](https://docs.nvidia.com/deeplearning/riva/user-guide/docs/quick-start-guide.html) - [Explore TTS Tutorials](https://docs.nvidia.com/deeplearning/riva/user-guide/docs/tutorials-tts-contents.html) ### Neural Machine Translation Integrate highly accurate text-to-text, speech-to-text, or speech-to-speech translation for up to 32 languages into your conversational application pipelines. - [Read NMT Documentation](https://docs.nvidia.com/deeplearning/riva/user-guide/docs/translation/translation-overview.html) - [Read Quick-Start Guide](https://docs.nvidia.com/deeplearning/riva/user-guide/docs/quick-start-guide.html) - [Explore NMT Tutorials](https://docs.nvidia.com/deeplearning/riva/user-guide/docs/tutorials-nmt-contents.html) * * * ## Nemotron Speech and NVIDIA Riva Learning Library The rows below are the raw result records for this component — one row per record. Column names are the source index fields; the full data is also downloadable in the links that follow. | links | title | excerpt | featured | products | sort_key | created_at | content_type | learning_levels | | --- | --- | --- | --- | --- | --- | --- | --- | --- | | {"url" => "https://catalog.ngc.nvidia.com/orgs/nvidia/teams/riva/models/punctuationcapitalization_en_us_bert_base", "text" => "Learn more"} | RIVA Punctuation and Capitalization for English | For each word in the input text, the model: 1) predicts a punctuation mark that should follow the word (if any), the model supports commas, periods and question marks) and 2) predicts if the word should be capitalized or not. | false | Riva | 2\|riva punctuation and capitalization for english | 1751575822 | Model | Advanced | | {"url" => "https://catalog.ngc.nvidia.com/orgs/nvidia/teams/riva/models/speechtotext_pt_br_conformer", "text" => "Learn more"} | RIVA Conformer ASR Brazilian Portuguese | Brazilian Portuguese (pt-BR) Conformer ASR model trained on ASR set 1.0 | false | Riva | 2\|riva conformer asr brazilian portuguese | 1743693798 | Model | Advanced | | {"url" => "https://catalog.ngc.nvidia.com/orgs/nvidia/teams/riva/models/speechtotext_en_us_conformer", "text" => "Learn more"} | RIVA Conformer ASR English - ASR set 6.0 | English Conformer ASR model trained on ASR set 6.0 | false | Riva | 2\|riva conformer asr english - asr set 6.0 | 1743692919 | Model | Advanced | | {"url" => "https://catalog.ngc.nvidia.com/orgs/nvidia/teams/riva/models/speechtotext_es_us_conformer", "text" => "Learn more"} | RIVA Conformer ASR Spanish - ASR set 3.0 | Spanish Conformer ASR model trained on ASR set 3.0 | false | Riva | 2\|riva conformer asr spanish - asr set 3.0 | 1743693488 | Model | Advanced | | {"url" => "https://catalog.ngc.nvidia.com/orgs/nvidia/teams/riva/models/speechtotext_en_us_citrinet", "text" => "Learn more"} | RIVA Citrinet ASR English en-US | English en-US Citrinet ASR model trained on RIVA ASR set | false | Riva | 2\|riva citrinet asr english en-us | 1680635131 | Model | Advanced | | {"url" => "https://catalog.ngc.nvidia.com/orgs/nvidia/teams/riva/models/speechtotext_ja_jp_conformer", "text" => "Learn more"} | RIVA Conformer ASR Japanese | Japanese Conformer ASR model trained on RIVA ASR set | false | Riva | 2\|riva conformer asr japanese | 1743693535 | Model | Advanced | | {"url" => "https://catalog.ngc.nvidia.com/orgs/nvidia/teams/riva/models/speechtotext_ru_ru_conformer", "text" => "Learn more"} | RIVA Conformer ASR Russian | Russian Conformer ASR model trained on RIVA ASR set | false | Riva | 2\|riva conformer asr russian | 1743693565 | Model | Advanced | | {"url" => "https://catalog.ngc.nvidia.com/orgs/nvidia/teams/riva/models/speechtotext_ar_ar_conformer", "text" => "Learn more"} | RIVA Conformer ASR Arabic | Arabic (ar-AR) Conformer ASR model trained on ASR set 2.0 | false | Riva | 2\|riva conformer asr arabic | 1743693189 | Model | Advanced | | {"url" => "https://catalog.ngc.nvidia.com/orgs/nvidia/teams/riva/models/speechtotext_hi_in_conformer", "text" => "Learn more"} | RIVA Conformer ASR Hindi | Hindi Conformer ASR model trained on RIVA ASR set | false | Riva | 2\|riva conformer asr hindi | 1743693434 | Model | Advanced | | {"url" => "https://catalog.ngc.nvidia.com/orgs/nvidia/teams/riva/models/speechtotext_en_us_lm", "text" => "Learn more"} | Riva ASR English LM | Base English n-gram LM trained on LibriSpeech, Switchboard and Fisher | false | Riva | 2\|riva asr english lm | 1719943458 | Model | Advanced | | {"url" => "https://catalog.ngc.nvidia.com/orgs/nvidia/teams/riva/containers/riva-speech", "text" => "Learn more"} | Riva Speech Skills | Riva Speech Skills is a scalable Conversational AI service platform. | false | Riva | 2\|riva speech skills | 1781627928 | Container | Intermediate | | {"url" => "https://developer.nvidia.com/blog/quickly-voice-your-apps-with-nvidia-nim-microservices-for-speech-and-translation/", "text" => "Learn more"} | Quickly Voice Your Apps with NVIDIA NIM Microservices for Speech and Translation | NVIDIA NIM, part of NVIDIA AI Enterprise, provides containers to self-host GPU-accelerated inferencing microservices for pretrained and customized AI models… | false | NIM, Riva | 2\|quickly voice your apps with nvidia nim microservices for speech and translation | 1726674523 | Blog | Intermediate, Advanced | | {"url" => "https://developer.nvidia.com/blog/how-to-take-a-rag-application-from-pilot-to-production-in-four-steps/", "text" => "Learn more"} | How to Take a RAG Application from Pilot to Production in Four Steps | Generative AI has the potential to transform every industry. Human workers are already using large language models (LLMs) to explain, reason about… | false | AI Enterprise, Metropolis, Morpheus, NeMo, NeMo Microservices, NeMo Retriever, NIM, RAPIDS, Riva | 2\|how to take a rag application from pilot to production in four steps | 1710774000 | Blog | Intermediate | | {"url" => "https://developer.nvidia.com/blog/spotlight-convai-reinvents-non-playable-character-interactions/", "text" => "Learn more"} | Spotlight: Convai Reinvents Non-Playable Character Interactions | Convai is a versatile developer platform for designing characters with advanced multimodal perception abilities. These characters are designed to integrate… | false | AI Enterprise, Avatar Cloud Engine (ACE), Audio2Face, NeMo, Omniverse, Riva, Triton Inference Server, Dynamo | 2\|spotlight: convai reinvents non-playable character interactions | 1704702600 | Blog | General Interest | | {"url" => "https://developer.nvidia.com/blog/nvidia-speech-and-translation-ai-models-set-records-for-speed-and-accuracy/", "text" => "Learn more"} | NVIDIA Speech and Translation AI Models Set Records for Speed and Accuracy | Speech and translation AI models developed at NVIDIA are pushing the boundaries of performance and innovation. The NVIDIA Parakeet automatic speech recognition… | false | LaunchPad, Riva | 2\|nvidia speech and translation ai models set records for speed and accuracy | 1710838800 | Blog | Intermediate | | {"url" => "https://developer.nvidia.com/blog/video-exploring-speech-ai-from-research-to-practical-production-applications/", "text" => "Learn more"} | Video: Exploring Speech AI from Research to Practical Production Applications | The integration of speech and translation AI into our daily lives is rapidly reshaping our interactions, from virtual assistants to call centers and augmented… | false | Riva | 2\|video: exploring speech ai from research to practical production applications | 1699344442 | Blog | General Interest | | {"url" => "https://developer.nvidia.com/blog/speech-ai-spotlight-visualizing-spoken-language-and-sounds-on-ar-glasses/", "text" => "Learn more"} | Speech AI Spotlight: Visualizing Spoken Language and Sounds on AR Glasses | Audio can include a wide range of sounds, from human speech to non-speech sounds like barking dogs and sirens. When designing accessible applications for people… | false | Riva | 2\|speech ai spotlight: visualizing spoken language and sounds on ar glasses | 1687507200 | Blog | General Interest | | {"url" => "https://nvda.ws/3VbKm5I#new_tab", "text" => "Learn more"} | Webinar: Empower Telco Contact Center Agents with Multi-Language Speech-AI-Customized Agent Assists | Join Infosys, NVIDIA and Quantiphi on June 7 to learn how to use speech and translation AI to improve agent assist solutions in multiple languages. | false | Riva | 2\|webinar: empower telco contact center agents with multi-language speech-ai-customized agent assists | 1684227600 | Blog | General Interest | | {"url" => "https://developer.nvidia.com/blog/unlocking-speech-ai-technology-for-global-language-users-top-qas/", "text" => "Learn more"} | Unlocking Speech AI Technology for Global Language Users: Top Q&As | This post summarizes the top questions asked during Unlocking Speech AI Technology for Global Language Users, a recorded talk from the Speech AI Summit 2022. | false | NeMo Framework, Riva | 2\|unlocking speech ai technology for global language users: top q&as | 1686045600 | Blog | General Interest | | {"url" => "https://developer.nvidia.com/blog/how-to-deploy-riva-speech-and-translation-ai-in-the-public-cloud/", "text" => "Learn more"} | How to Deploy NVIDIA Riva Speech and Translation AI in the Public Cloud | From start-ups to large enterprises, businesses use cloud marketplaces to find the new solutions needed to quickly transform their businesses. | false | Riva | 2\|how to deploy nvidia riva speech and translation ai in the public cloud | 1693303200 | Blog | Intermediate | | {"url" => "https://nvda.ws/48oZgfp#new_tab", "text" => "Learn more"} | Workshop: Building Conversational AI Applications | Learn how to build and deploy production-quality conversational AI apps with real-time transcription and NLP. | false | NeMo Framework, Riva | 2\|workshop: building conversational ai applications | 1695204000 | Blog | Intermediate | | {"url" => "https://developer.nvidia.com/blog/how-speech-recognition-improves-customer-service-in-telecommunications/", "text" => "Learn more"} | How Speech Recognition Improves Customer Service in Telecommunications | The telecommunication industry has seen a proliferation of AI-powered technologies in recent years, with speech recognition and translation leading the charge. | false | NeMo Framework, Riva | 2\|how speech recognition improves customer service in telecommunications | 1683018000 | Blog | Intermediate | | {"url" => "https://developer.nvidia.com/blog/speech-ai-spotlight-how-pendulum-nabs-harmful-narratives-online/", "text" => "Learn more"} | Speech AI Spotlight: How Pendulum Nabs Harmful Narratives Online | Over 55% of the global population uses social media, easily sharing online content with just one click. While connecting with others and consuming entertaining… | false | Riva, Triton Inference Server, Dynamo | 2\|speech ai spotlight: how pendulum nabs harmful narratives online | 1675846800 | Blog | Intermediate | | {"url" => "https://developer.nvidia.com/blog/create-xr-experiences-using-natural-language-voice-commands-test-project-mellon/", "text" => "Learn more"} | Create XR Experiences Using Natural-Language Voice Commands: Test Project Mellon | At GTC 2023, NVIDIA announced that developers can start testing Project Mellon to explore creating hands-free extended reality (XR) experiences controlled by... | false | Riva | 2\|create xr experiences using natural-language voice commands: test project mellon | 1679558400 | Blog | | | {"url" => "https://nvda.ws/3EOo4QR#new_tab", "text" => "Learn more"} | Top Video Streaming and Conferencing Sessions at NVIDIA GTC 2023 | Learn about advancements in video conferencing that have transformed how we communicate. | false | Maxine, Riva | 2\|top video streaming and conferencing sessions at nvidia gtc 2023 | 1677834000 | Blog | General Interest | | {"url" => "https://nvda.ws/3onruEu#new_tab", "text" => "Learn more"} | Workshop: How to Enable Your Product with Voice Interface | This hands-on workshop will guide you through the process of voice-enabling your product, from familiarizing yourself with NVIDIA Riva to assessing the costs and... | false | Riva | 2\|workshop: how to enable your product with voice interface | 1681896548 | Blog | Beginner | | {"url" => "https://nvda.ws/3LCd3po#new_tab", "text" => "Learn more"} | Webinar: How Telcos Transform Customer Experiences with Conversational AI | Join Infosys, Quantiphi, Talkmap, and NVIDIA on May 31 for a live webinar to learn how telecommunications companies are using AI to improve operational efficiency and... | false | Riva | 2\|webinar: how telcos transform customer experiences with conversational ai | 1683018000 | Blog | | | {"url" => "https://www.nvidia.com/en-us/launchpad/ai/intelligent-virtual-assistant/#new_tab", "text" => "Learn more"} | New Hands-on Lab: Intelligent Virtual Assistant | Learn to build an engaging and intelligent virtual assistant with NVIDIA AI workflows powered by NVIDIA Riva in this free hands-on lab from NVIDIA LaunchPad… | false | LaunchPad, Riva | 2\|new hands-on lab: intelligent virtual assistant | 1674565200 | Blog | | | {"url" => "https://nvda.ws/448NHGO#new_tab", "text" => "Learn more"} | Explainer: What Is Agent Assist? | Agent-assist technology uses AI and ML to provide facts and make real-time suggestions that help human agents across retail, telecom, and other industries conduct... | false | Riva | 2\|explainer: what is agent assist? | 1683198000 | Blog | General Interest | | {"url" => "https://nvda.ws/3jRS331#new_tab", "text" => "Learn more"} | Top Speech AI Developer Day Sessions at NVIDIA GTC 2023 | Explore the latest advances in accurate and customizable automatic speech recognition, multi-language translation, and text-to-speech. | false | Riva | 2\|top speech ai developer day sessions at nvidia gtc 2023 | 1676383200 | Blog | | | {"url" => "https://nvda.ws/41sYdaR#new_tab", "text" => "Learn more"} | Top Conversational AI Sessions at NVIDIA GTC 2023 | Learn about the latest tools, trends, and technologies for building and deploying conversational AI. | false | NeMo Framework, Riva | 2\|top conversational ai sessions at nvidia gtc 2023 | 1677583827 | Blog | | | {"url" => "https://developer.nvidia.com/blog/autoscaling-nvidia-riva-deployment-with-kubernetes-for-speech-ai-in-production/", "text" => "Learn more"} | Autoscaling NVIDIA Riva Deployment with Kubernetes for Speech AI in Production | Learn how to deploy NVIDIA Riva servers on a large scale with Kubernetes for autoscaling and Traefik for load balancing. | false | Riva, Triton Inference Server, Dynamo | 2\|autoscaling nvidia riva deployment with kubernetes for speech ai in production | 1673515800 | Blog | Intermediate | | {"url" => "https://developer.nvidia.com/blog/new-languages-enhanced-cybersecurity-and-medical-ai-frameworks-unveiled-at-gtc/", "text" => "Learn more"} | New Languages, Enhanced Cybersecurity, and Medical AI Frameworks Unveiled at GTC | Major updates to AI frameworks for building real-time speech AI apps, designing high-performing recommenders at scale, applying AI to cybersecurity challenges, and... | false | BioNeMo, Clara Holoscan, cuOpt, DRIVE, Isaac SDK, Maxine, Merlin, Metropolis, MONAI, Morpheus, NeMo Framework, Riva, Triton Inference Server, Dynamo | 2\|new languages, enhanced cybersecurity, and medical ai frameworks unveiled at gtc | 1663748280 | Blog | | | {"url" => "https://developer.nvidia.com/blog/developing-the-next-generation-of-extended-reality-applications-with-speech-ai/", "text" => "Learn more"} | Developing the Next Generation of Extended Reality Applications with Speech AI | Technical tutorial for extended reality (XR) developers adding speech AI skills, like ASR and TTS, to AR/VR/MR applications using Riva SDK as an example. | false | CloudXR, Riva | 2\|developing the next generation of extended reality applications with speech ai | 1663146000 | Blog | | | {"url" => "https://developer.nvidia.com/blog/reducing-development-time-for-intelligent-virtual-assistants-in-contact-centers/", "text" => "Learn more"} | Reducing Development Time for Intelligent Virtual Assistants in Contact Centers | As the global service economy grows, companies rely increasingly on contact centers to drive better customer experiences, increase customer satisfaction… | false | Riva | 2\|reducing development time for intelligent virtual assistants in contact centers | 1671091200 | Blog | | | {"url" => "https://info.nvidia.com/ai-virtual-assistant-for-finance-contact-centers-webinar.html#new_tab", "text" => "Learn more"} | Upcoming Webinar: Building an Intelligent Virtual Assistant for Financial Services | Join this webinar on January 25 and learn how to build a voice-enabled intelligent virtual assistant to improve customer experiences at contact centers. | false | Riva | 2\|upcoming webinar: building an intelligent virtual assistant for financial services | 1673427600 | Blog | | | {"url" => "https://developer.nvidia.com/blog/making-an-nvidia-riva-asr-service-for-a-new-language/", "text" => "Learn more"} | Making an NVIDIA Riva ASR Service for a New Language | NVIDIA Riva facilitates the process of creating ASR services with the tools and methodologies to help you realize your skills, all the way from raw data to a... | false | Riva | 2\|making an nvidia riva asr service for a new language | 1666951200 | Blog | Intermediate | | {"url" => "https://courses.nvidia.com/courses/course-v1:DLI+S-FX-04+V1/?nvid=nv-int-tblg-973505#cid=dli03_nv-int-tblg_en-us#new_tab", "text" => "Learn more"} | New Course: Get Started with Highly Accurate Custom ASR for Speech AI | Learn how to build, train, customize, and deploy a GPU-accelerated automatic speech recognition service with NVIDIA Riva in this self-paced course. | false | Riva | 2\|new course: get started with highly accurate custom asr for speech ai | 1666603800 | Blog | | | {"url" => "https://developer.nvidia.com/blog/speech-ai-technology-enables-natural-interactions-with-service-robots/", "text" => "Learn more"} | Speech AI Technology Enables Natural Interactions with Service Robots | From taking your order and serving you food in a restaurant to playing poker with you, service robots are becoming increasingly prevalent. Globally… | false | Riva | 2\|speech ai technology enables natural interactions with service robots | 1671207787 | Blog | | | {"url" => "https://developer.nvidia.com/blog/solving-automatic-speech-recognition-deployment-challenges/", "text" => "Learn more"} | Solving Automatic Speech Recognition Deployment Challenges | Top speech recognition challenges developers face when building and deploying an ASR application and how to solve them using a speech AI-optimized workflow. | false | Riva | 2\|solving automatic speech recognition deployment challenges | 1661936400 | Blog | | | {"url" => "https://developer.nvidia.com/riva/getting-started#new_tab", "text" => "Learn more"} | Just Released: New Updates to NVIDIA Riva | Build better GPU-accelerated Speech AI applications with the latest NVIDIA Riva updates, including enterprise support. | false | Riva | 2\|just released: new updates to nvidia riva | 1664186400 | Blog | | | {"url" => "https://developer.nvidia.com/blog/deploying-nvidia-riva-multilingual-asr-with-whisper-and-canary-architectures-while-selectively-deactivating-nmt/", "text" => "Learn more"} | Deploying NVIDIA Riva Multilingual ASR with Whisper and Canary Architectures While Selectively Deactivating NMT | NVIDIA has consistently developed automatic speech recognition (ASR) models that set the benchmark in the industry. Earlier versions of NVIDIA Riva… | false | NIM, Riva | 2\|deploying nvidia riva multilingual asr with whisper and canary architectures while selectively deactivating nmt | 1740048888 | Blog | Intermediate | | {"url" => "https://developer.nvidia.com/blog/spotlight-personal-ai-brings-ai-receptionists-to-small-business-owners-with-nvidia-riva/", "text" => "Learn more"} | Spotlight: Personal AI Brings AI Receptionists to Small Business Owners with NVIDIA Riva | It’s 10 p.m. on a Tuesday when the phone rings at the Sapochnick Law Firm, a specialized law practice in San Diego, California. The caller, a client of the firm… | false | NIM, Riva | 2\|spotlight: personal ai brings ai receptionists to small business owners with nvidia riva | 1745941447 | Blog | Beginner | | {"url" => "https://developer.nvidia.com/blog/expanding-ai-agent-interface-options-with-2d-and-3d-digital-human-avatars/", "text" => "Learn more"} | Expanding AI Agent Interface Options with 2D and 3D Digital Human Avatars | When interfacing with generative AI applications, users have multiple communication options—text, voice, or through digital avatars. | false | Avatar Cloud Engine (ACE), Audio2Face, NeMo Microservices, NeMo Retriever, NIM, NIM Agent Blueprint, Omniverse, Riva | 2\|expanding ai agent interface options with 2d and 3d digital human avatars | 1731516803 | Blog | Intermediate | | {"url" => "https://developer.nvidia.com/blog/build-a-digital-human-interface-for-ai-apps-with-an-nvidia-nim-agent-blueprint/", "text" => "Learn more"} | Build a Digital Human Interface for AI Apps with an NVIDIA NIM Agent Blueprint | Providing customers with quality service remains a top priority for businesses across industries, from answering questions and troubleshooting issues to… | false | AI Enterprise, Avatar Cloud Engine (ACE), NIM, Riva | 2\|build a digital human interface for ai apps with an nvidia nim agent blueprint | 1727271000 | Blog | Intermediate | | {"url" => "https://developer.nvidia.com/blog/enhancing-multilingual-human-like-speech-and-voice-cloning-with-nvidia-riva-tts/", "text" => "Learn more"} | Enhancing Multilingual Human-Like Speech and Voice Cloning with NVIDIA Riva TTS | While speech AI is used to build digital assistants and voice agents, its impact extends far beyond these applications. Core technologies like text-to-speech… | false | AI Enterprise, NIM, Riva | 2\|enhancing multilingual human-like speech and voice cloning with nvidia riva tts | 1752485400 | Blog | Intermediate | | {"url" => "https://developer.nvidia.com/blog/an-easy-introduction-to-multimodal-retrieval-augmented-generation-for-video-and-audio/", "text" => "Learn more"} | An Easy Introduction to Multimodal Retrieval-Augmented Generation for Video and Audio | Building a multimodal retrieval-augmented generation (RAG) system is challenging. The difficulty comes from capturing and indexing information from across… | false | Riva | 2\|an easy introduction to multimodal retrieval-augmented generation for video and audio | 1734339600 | Blog | Intermediate | | {"url" => "https://developer.nvidia.com/blog/nvidia-speech-ai-models-deliver-industry-leading-accuracy-and-performance/", "text" => "Learn more"} | NVIDIA Speech AI Models Deliver Industry-Leading Accuracy and Performance | NVIDIA is driving state-of-the-art performance, efficiency, and accessibility in both speech AI and language models, setting the stage for innovations that are… | false | NIM, Riva | 2\|nvidia speech ai models deliver industry-leading accuracy and performance | 1749032191 | Blog | Intermediate | | {"url" => "https://huggingface.co/spaces/hf-audio/open_asr_leaderboard#new_tab", "text" => "Learn more"} | NVIDIA Canary‑Qwen‑2.5B: Open‑Source ASR/LLM for Superior Transcription and Summarization | Top‑ranked on the HuggingFace Open‑ASR leaderboard, the model is production‑ready. | false | Riva | 2\|nvidiacanary-qwen-2.5b: open-source asr/llm for superior transcription and summarization | 1752747447 | Blog | Beginner | | {"url" => "https://developer.nvidia.com/blog/building-a-speech-enabled-ai-virtual-assistant-with-riva-on-amazon-ec2/", "text" => "Learn more"} | Building a Speech-Enabled AI Virtual Assistant with NVIDIA Riva on Amazon EC2 | Learn how to get started with NVIDIA Riva, a fully accelerated speech AI SDK, on AWS EC2 using Jupyter Notebooks and a sample virtual assistant application. | false | Riva | 2\|building a speech-enabled ai virtual assistant with nvidia riva on amazon ec2 | 1658997000 | Blog | | | {"url" => "https://www.minervacq.com/minervacq-first-to-deploy-nvidia-riva-in-energy#new_tab", "text" => "Learn more"} | Minerva CQ Deploys NVIDIA Riva Enterprise in the Energy Sector | Learn how NVIDIA Inception member Minerva CQ is using NVIDIA Riva to deliver faster, personalized experiences within a global EV charging and electric mobility company. | false | Riva | 2\|minerva cq deploys nvidia riva enterprise in the energy sector | 1656597600 | Blog | | | {"url" => "https://developer.nvidia.com/blog/identify-speakers-in-meetings-calls-and-voice-apps-in-real-time-with-nvidia-streaming-sortformer/", "text" => "Learn more"} | Identify Speakers in Meetings, Calls, and Voice Apps in Real-Time with NVIDIA Streaming Sortformer | September 2026 update: Our latest diarization release is Nemotron 3 Diarization. It’s #1 on the VoiceArena leaderboard, and supports up to 8 speakers. | false | NeMo, Riva | 2\|identify speakers in meetings, calls, and voice apps in real-time with nvidia streaming sortformer | 1755507600 | Blog | Intermediate | | {"url" => "https://developer.nvidia.com/blog/how-language-neutralization-is-transforming-customer-service-contact-centers/", "text" => "Learn more"} | How Language Neutralization Is Transforming Customer Service Contact Centers | The powerful language neutralization offered by Infosys Cortex and based on NVIDIA Riva speech and translation enables contact center agents to communicate effectively... | false | Riva | 2\|how language neutralization is transforming customer service contact centers | 1685462314 | Blog | Intermediate | | {"url" => "https://developer.nvidia.com/blog/enhancing-customer-experience-in-telecom-with-nvidia-customized-speech-ai/", "text" => "Learn more"} | Enhancing Customer Experience in Telecom with NVIDIA Customized Speech AI | Learn why conversational AI systems are essential and why it is important to have a high level of transcription accuracy for optimal performance in downstream tasks. | false | Riva | 2\|enhancing customer experience in telecom with nvidia customized speech ai | 1685433600 | Blog | Intermediate | | {"url" => "https://developer.nvidia.com/blog/boost-meeting-productivity-with-ai-powered-note-taking-and-summarization/", "text" => "Learn more"} | Boost Meeting Productivity with AI-Powered Note-Taking and Summarization | Meetings are the lifeblood of an organization. They foster collaboration and informed decision-making. They eliminate silos through brainstorming and problem… | false | Riva | 2\|boost meeting productivity with ai-powered note-taking and summarization | 1701262800 | Blog | General Interest | | {"url" => "https://developer.nvidia.com/blog/introducing-riva-a-gpu-accelerated-sdk-for-developing-speech-ai-applications/", "text" => "Learn more"} | Introducing NVIDIA Riva: A GPU-Accelerated SDK for Developing Speech AI Applications | Learn about the Riva SDK and its use in developing speech AI applications. We also discuss Riva’s architecture, key features, and components for building speech AI... | false | Riva, TensorRT, Triton Inference Server, Dynamo | 2\|introducing nvidia riva: a gpu-accelerated sdk for developing speech ai applications | 1670513839 | Blog | | | {"url" => "https://catalog.ngc.nvidia.com/orgs/nvidia/teams/riva/models/parakeet-rnnt-riva-1-1b-unified-ml-cs-universal", "text" => "Learn more"} | RIVA Parakeet-RNNT-XXL-1.1B ASR Multilingual Universal Tokenizer | Multilingual Parakeet-RNNT-XXL-1.1B ASR model with Universal Tokenizer trained on ASR set 1.0 | false | Riva | 2\|riva parakeet-rnnt-xxl-1.1b asr multilingual universal tokenizer | 1752784810 | Model | Advanced | | {"url" => "https://developer.nvidia.com/blog/enhancing-phone-customer-service-with-asr-customization/", "text" => "Learn more"} | Enhancing Phone Customer Service with ASR Customization | At the core of understanding people correctly and having natural conversations is automatic speech recognition (ASR). To make customer-led voice assistants and… | false | Riva | 2\|enhancing phone customer service with asr customization | 1704790800 | Blog | General Interest | | {"url" => "https://developer.nvidia.com/blog/build-real-time-multimodal-xr-apps-with-nvidia-ai-blueprint-for-video-search-and-summarization/", "text" => "Learn more"} | Vision Language Model Prompt Engineering Guide for Image and Video Understanding | | false | Metropolis, Isaac Sim, NIM, Omniverse, Riva | 1\|vision language model prompt engineering guide for image and video understanding | 1791277765 | Blog | | | {"url" => "https://catalog.ngc.nvidia.com/orgs/nvidia/teams/riva/models/speechsynthesis_multilingual_magpietts_ipa", "text" => "Learn more"} | RIVA Magpie-TTS Multilingual | Riva NeMo-MagpieTTS Multilingual IPA multispeaker model with Emotions | false | Riva | 2\|riva magpie-tts multilingual | 1752784739 | Model | Advanced | | {"url" => "https://catalog.ngc.nvidia.com/orgs/nvidia/teams/riva/models/sortformer_diarizer", "text" => "Learn more"} | Sortformer Speaker Diarization | Sortformer based speaker diarization model | false | Riva | 2\|sortformer speaker diarization | 1752784762 | Model | Advanced | | {"url" => "https://developer.nvidia.com/blog/advance-video-analytics-ai-agents-using-the-nvidia-ai-blueprint-for-video-search-and-summarization/", "text" => "Learn more"} | Build Real-Time Multimodal XR Apps with NVIDIA AI Blueprint for Video Search and Summarization | | false | Riva, Metropolis | 1\|build real-time multimodal xr apps with nvidia ai blueprint for video search and summarization | 1791277765 | Blog | | | {"url" => "https://developer.nvidia.com/blog/maximize-robotics-performance-by-post-training-nvidia-cosmos-reason/", "text" => "Learn more"} | Advance Video Analytics AI Agents Using the NVIDIA AI Blueprint for Video Search and Summarization | | false | Metropolis, Cosmos | 1\|advance video analytics ai agents using the nvidia ai blueprint for video search and summarization | 1791277765 | Blog | | | {"url" => "https://developer.nvidia.com/blog/how-to-integrate-computer-vision-pipelines-with-generative-ai-and-reasoning/", "text" => "Learn more"} | Build a Real-Time Visual Inspection Pipeline with NVIDIA TAO 6 and NVIDIA DeepStream 8 | | false | Metropolis, Blueprint, Cosmos, NIM | 1\|build a real-time visual inspection pipeline with nvidia tao 6 and nvidia deepstream 8 | 1791277765 | Blog | | | {"url" => "https://developer.nvidia.com/blog/optimizing-semiconductor-defect-classification-with-generative-ai-and-vision-foundation-models/", "text" => "Learn more"} | How to build Visual AI Agents with NVIDIA Cosmos Reason and Metropolis | | false | Metropolis, Cosmos, TAO Toolkit | 1\|how to build visual ai agents with nvidia cosmos reason and metropolis | 1791277765 | Video | | | {"url" => "https://developer.nvidia.com/blog/low-code-building-blocks-for-speech-ai-robotics/", "text" => "Learn more"} | Low-Code Building Blocks for Speech AI Robotics | Learn how to add speech AI skills, like speech-to-text (STT) and text-to-speech (TTS), to your intelligent robotics application using low-code building blocks. | false | Jetson Orin, Riva | 2\|low-code building blocks for speech ai robotics | 1663846380 | Blog | | | {"url" => "https://developer.nvidia.com/blog/vision-language-model-prompt-engineering-guide-for-image-and-video-understanding/", "text" => "Learn more"} | Build an Agentic Video Workflow with Video Search and Summarization | | false | Riva, Metropolis | 1\|build an agentic video workflow with video search and summarization | 1791277765 | Blog | | | {"url" => "https://developer.nvidia.com/blog/build-a-real-time-visual-inspection-pipeline-with-nvidia-tao-6-and-nvidia-deepstream-8/", "text" => "Learn more"} | Maximize Robotics Performance by Post-Training NVIDIA Cosmos Reason | | false | Metropolis | 1\|maximize robotics performance by post-training nvidia cosmos reason | 1791277765 | Blog | | | {"url" => "", "text" => "Learn more"} | Optimizing Semiconductor Defect Classification with Generative AI and Vision Foundation Models | | false | TAO Toolkit, Metropolis | 1\|optimizing semiconductor defect classification with generative ai and vision foundation models | 1791277765 | Blog | | | {"url" => "https://developer.nvidia.com/blog/build-an-agentic-video-workflow-with-video-search-and-summarization/", "text" => "Learn more"} | Build Multimodal Visual AI Agents Powered by NVIDIA NIM | | false | Metropolis, Morpheus, NIM | 1\|build multimodal visual ai agents powered by nvidia nim | 1791277765 | Blog | | | {"url" => "https://www.youtube.com/watch?v=E0c4vTfeyqU", "text" => "Learn more"} | How to Integrate Computer Vision Pipelines with Generative AI and Reasoning | | false | Metropolis | 1\|how to integrate computer vision pipelines with generative ai and reasoning | 1791277765 | Blog | | | {"url" => "https://developer.nvidia.com/blog/building-the-ai-grid-with-nvidia-orchestrating-intelligence-everywhere/", "text" => "Learn more"} | Building the AI Grid with NVIDIA: Orchestrating Intelligence Everywhere  | AI-native services are exposing a new bottleneck in AI infrastructure: As millions of users, agents, and devices demand access to intelligence… | false | Holoscan, Maxine, Metropolis, Riva | 2\|building the ai grid with nvidia: orchestrating intelligence everywhere | 1773742400 | Blog | Intermediate | | {"url" => "https://developer.nvidia.com/blog/build-speech-ai-in-multiple-languages-and-train-large-language-models-with-the-latest-from-riva-and-nemo-megatron/", "text" => "Learn more"} | Build Speech AI in Multiple Languages and Train Large Language Models with the Latest from Riva and NeMo Framework | Read a recap of conversational AI announcements from NVIDIA GTC. | false | NeMo Framework, Riva | 2\|build speech ai in multiple languages and train large language models with the latest from riva and nemo framework | 1648458000 | Blog | | | {"url" => "https://catalog.ngc.nvidia.com/orgs/nvidia/teams/riva/models/speechtotext_ar_ar_lm", "text" => "Learn more"} | Riva ASR Arabic LM | Base Arabic 4-gram LM | false | Riva | 2\|riva asr arabic lm | 1681842221 | Model | Advanced | [Download the raw results data (JSON)](https://developer.nvidia.com/search-data/ll_riva.json) * * * ## More Resources ![A decorative image representing Developer Community](https://developer.download.nvidia.com/icons/m48-people-group.svg) ### Explore the Community ![img-alt-text](https://developer.download.nvidia.com/icons/m48-certification-ribbon-2.svg) ### Get Training and Certification ![A decorative image representing Inception for Startups](https://developer.download.nvidia.com/icons/m48-ai-startup.svg) ### Meet the Program for Startups * * * ## Ethical AI NVIDIA’s platforms and application frameworks enable developers to build a wide array of AI applications. Always consider potential algorithmic bias when choosing or creating the models being deployed. Work with the model’s developer to ensure that it meets the requirements for the relevant industry and use case; that the necessary instructions and documentation are provided to understand error rates, confidence intervals, and results; and that the model is being used under the conditions and in the manner intended. Stay up to date on the latest speech and translation AI news from NVIDIA. [Sign Up](https://www.nvidia.com/en-us/deep-learning-ai/solutions/conversational-ai/speech-ai-news/)