NVIDIA TTS NIM Microservice API References#
The TTS NIM microservice exposes three APIs for speech synthesis: an HTTP REST API for simple offline and streaming synthesis, a gRPC API based on protocol buffers for batch and streaming synthesis, and a WebSocket-based realtime API for low-latency interactive use cases such as conversational agents.
HTTP REST API
REST endpoints for voice listing, offline synthesis, and streaming synthesis. Simple multipart/form-data requests with no gRPC dependency.
gRPC Protobuf API
Protocol buffer definitions for the RivaSpeechSynthesis gRPC service, including batch and streaming text-to-speech methods.
Realtime WebSocket API
WebSocket event-based API reference for real-time text streaming and speech synthesis.