Getting Started

• Overview
• Quick Start Guide
• Riva Enterprise
• Release Notes

Tutorials

• Speech Recognition
  • How do I use Riva ASR APIs with out-of-the-box models?
    • NVIDIA Riva Overview
    • Transcription with Riva ASR APIs
    • Go deeper into Riva capabilities
  • How to Improve Recognition of Specific Words
    • Overview of Riva customization techniques
    • 1. Word boosting
    • 2. Custom vocabulary
    • 3. Custom pronunciation (Lexicon mapping)
    • 4. Retrain language model
    • 5. Fine tune the acoustic model
  • Conclusion
  • How do I boost specific words at runtime with word boosting?
    • NVIDIA Riva Overview
    • Word boosting with Riva ASR APIs
    • Go deeper into Riva capabilities
  • How to Customize Riva ASR Vocabulary and Pronunciation with Lexicon Mapping
    • Overview
    • What can be customized?
    • Extending the vocabulary
    • Customizing pronunciation with lexicon mapping
    • Go deeper into Riva capabilities
  • How to pretrain a Riva ASR Language Modeling (n-gram) with TAO Toolkit
    • NVIDIA Riva Overview
    • TAO Toolkit
    • Language Modeling
    • Let’s Dig in: Riva Language Modeling using TAO
    • TAO Toolkit workflow
  • How to fine-tune a Riva ASR Acoustic Model (Citrinet) with TAO Toolkit
    • NVIDIA Riva Overview
    • Train Adapt Optimize (TAO) Toolkit
    • Automatic Speech Recognition (ASR)
    • ASR using TAO
    • What’s Next?
  • How to deploy custom Acoustic Model (Citrinet) trained with TAO Toolkit on Riva
    • NVIDIA Riva Overview
    • Train, Adapt, and Optimize TAO Toolkit
    • Prerequisites
    • Riva ServiceMaker
    • Start the Riva Server
    • Run Inference
• Speech Recognition - New Language Adaptation
  • The Making of RIVA German ASR Service
    • Overview
    • 1. Data collection
    • 2. Data preparation
    • 3. Training and validation
    • 4. Riva deployment
  • Conclusion
  • The Making of RIVA Hindi ASR Service
    • Overview
    • 1. Data collection
    • 2. Data preparation
    • 3. Training and validation
    • 4. Model Testing
    • 5. Riva Deployment
    • Conclusion
  • The Making of the Riva Mandarin ASR Service
    • Overview
    • 1. Data collection
    • 2. Data preparation
    • 3. Training and validation
    • 4. Riva deployment
  • Conclusion
• Speech Synthesis
  • How do I use Riva TTS APIs with out-of-the-box models?
    • NVIDIA Riva Overview
    • Speech generation with Riva TTS APIs
    • Go deeper into Riva capabilities
  • How do I customize Riva TTS audio output with SSML?
    • NVIDIA Riva Overview
    • Customizing Riva TTS audio output with SSML
    • Go deeper into Riva capabilities
  • How to train Riva TTS models (FastPitch and HiFiGAN) with TAO Toolkit
    • NVIDIA Riva Overview
    • NVIDIA TAO Toolkit Overview
    • Text to Speech
    • TTS using TAO
    • TTS Inference with TAO Toolkit
    • What’s Next?
  • How to Deploy a custom TTS Models (FastPitch and HiFi-GAN) trained with TAO Toolkit Riva
    • NVIDIA Riva Overview
    • Train, Adapt and Optimize (TAO) Toolkit
    • Speech generation with Riva TTS APIs
    • Riva ServiceMaker
    • Start the Riva Server
    • Run Inference

Architecture

• Overview
  • Using Your Own Data
  • Model Development with TAO Toolkit
  • Model Development with NeMo
  • Riva Build
  • Riva Deploy
  • Deploying Your Custom Model into Riva
  • Build Process
  • Deploy Process
• Clients in a New Programming Language
  • Download Riva Skills API
  • Generate Language Bindings
  • Write the Client

Speech Recognition

• ASR Overview
  • Try It Out
  • Language Support
  • Features
• ASR Customization Best Practices
  • Choosing a Customization Technique
  • Model Adaptation
• Pipeline Configuration
  • Streaming/Offline Recognition
  • Language Models
  • Flashlight Decoder Lexicon
  • Beginning/End of Utterance Detection
  • Generating Multiple Transcript Hypotheses
  • (Advanced) Sharing Acoustic and Feature Extractor Models Across Multiple ASR Pipelines
  • Riva-build Optional Parameters
• Performance
  • Evaluation Process
  • Results
  • Hardware Specifications

Speech Synthesis

• TTS Overview
  • Try It Out
  • Voices
• Controllable TTS (SSML)
  • Prosody
  • Phoneme
  • Try It Now
• Custom Models
  • Model Deployment
  • Creating Riva Files
  • Custom Pronunciations
  • Multispeaker Models
  • Custom Voice
  • Pretrained Models
  • Pipeline Configuration
  • Pretrained Quick Start Pipelines
• Performance
  • Evaluation Process
  • Results
  • Hardware Specifications
  • Performance Considerations

Natural Language Processing

• NLP Overview
• Custom Models
  • Pretrained Models

Installation

• Best Practices
  • Security
  • Autoscaling Configurations
  • Monitoring GPU and CPU Telemetry
  • Load-Balancing Types
  • Optimizing Resource Usage for Embedded Platforms
• Local (Docker)
  • Configuring
  • Downloading Required Models and Containers from NGC
  • Launching the Servers and Client Container
  • Stopping
  • Clean-up
• Kubernetes
  • Installation with Helm
  • Kubernetes Secrets
  • Riva Settings
  • Ingress Controller
  • Load Balancer
  • Autoscaling Configurations
  • gRPC Streams with SSL/TLS
  • Monitoring GPU and CPU Telemetry
  • Load-Balancing Types
• How to Deploy Riva at Scale on AWS with EKS
  • Prerequisites
  • Creating the EKS Cluster
  • Deploying the Riva API
  • Deploying the Traefik edge router
  • Creating the IngressRoute
  • Deploying a Sample Client
  • Scaling the cluster
• NVIDIA Fleet Command
  • Application Setup
  • Deployment Setup

SDKs and Sample Apps

• Python
  • Installation
  • Install the Python Wheel
  • Python Example
• Command-line Clients
  • Speech Recognition
  • Natural Language Processing
  • Speech Synthesis
• Sample Apps
  • Riva Contact
  • Riva Virtual Assistant Example
  • Virtual Assistant (with Rasa)
  • Virtual Assistant (with Google Dialogflow)
  • SpeechSquad
  • AudioCodes VoiceGateway Sample

Reference

• Models
  • Speech Recognition
  • Natural Language Processing
  • Speech Synthesis
• gRPC & Protocol Buffers
  • riva/proto/riva_asr.proto
  • riva/proto/riva_nlp.proto
  • riva/proto/riva_tts.proto
  • riva/proto/riva_audio.proto
  • riva/proto/health.proto
• Troubleshooting
  • NGC
  • Model Export and ServiceMaker
  • Services
  • Client Integration
  • Riva Helm Deployment
  • SpeechSquad
• Support Matrix
  • Riva Speech Skills 2.2.0
  • Riva Speech Skills 2.1.0
  • Riva Speech Skills 2.0.0
  • Riva Speech Skills 1.10.0 Beta
  • Riva Speech Skills 1.9.0 Beta
  • Riva Speech Skills 1.8.0 Beta
  • Riva Speech Skills 1.7.0 Beta
  • Riva Speech Skills 1.6.0 Beta
  • Riva Speech Skills 1.5.0 Beta
  • Riva Speech Skills 1.4.0 Beta
• Archives
• Upgrading
• Acknowledgements
  • Google APIs
  • GoogleTest
  • gflags
  • Google Logging Library (glog)
  • speexdsp
  • libFLAC
  • gRPC
  • Triton Inference Server
  • NVlabs cub
  • KenLM
  • Kaldi
  • grpc_health_probe
  • gRPCurl
  • OpenFST
  • Yamale
  • PyTorch
  • requests
  • PyCUDA
  • RapidJSON
  • protobuf
  • onnx
  • librosa
  • omegaconf
  • utf8proc
  • re2
  • thrax
  • Sparrowhawk
  • SentencePiece
  • YouTokenToMe
• Notice
  • Google
  • Trademarks
  • Copyright