Documentation
User and developer documentation for NeMo-Speech.cpp. The shortest path to a transcript lives in the
; everything below is the detailed manual.
Start with:
HTTP/realtime and optional gRPC server
ASR
- ready-to-run GGUF downloads, custom checkpoint conversion, and quantization options.
- full asr.* key reference plus the decoding / word-boosting / VAD / endpointing / diarization / postprocessing knobs.
Feature matrix and customization
- which models support which features, and request-time vs startup customization.
TTS
- ready-to-run MagpieTTS and NanoCodec GGUF downloads, tokenizer setup, and custom conversion.
- tts.* key reference and serving.
NMT
- obtain and convert Riva-Translate models for the llama.cpp runtime.
- nmt.* key reference and serving.
S2S
- build and serve the streaming VoiceChat pipeline.
- convert complete model repositories.
- build and run the VoiceChat server image.
- s2s.* settings and runtime tuning.
- streaming, tool responses, and concurrency.
Developer guide
- implementation and performance internals.
- check_backend_coverage.
- neural microbatching and streaming-state arenas.
- the project-specific ggml changes.
- the in-tree drop-in cuBLAS replacement.