v0.1.5 v0.1.0
m-bain/whisperXAug 18, 2026 → Aug 19, 20261 day between releases
What's New
Newer release is a different software product (NeMo-Speech.cpp) compared to the older release (Open-SEO).
New Features
- Added speech recognition support for Nemotron 3.5 ASR and Parakeet TDT
- Integrated speaker diarization using Streaming Sortformer v2
- Enabled speech synthesis using MagpieTTS and NanoCodec
- Added text and speech translation capabilities with Riva Translate
- Included utilities for VAD, endpointing, punctuation, and capitalization
Improvements
- Introduced a unified CLI for model discovery, downloads, and caching
- Added an OpenAI-compatible HTTP API and realtime WebSocket transcription
- Provided native C API integration and cross-platform backend support
- Enabled CPU, CUDA, Vulkan, and Metal backend support
Upgrade Notes
The newer release is a completely different software product (NeMo-Speech.cpp) compared to the older release (Open-SEO). No upgrade path exists between these repositories.
The Two Releases
v0.1.5
Aug 18, 2026· 2 months ago
Adds flexibility to research query types by introducing 'subfolder' selection and expands Local SEO capabilities with new MCP tools for managing business profiles and reviews.
Full release notesv0.1.0
Aug 19, 2026· 2 months ago
This is the first versioned release of NeMo-Speech.cpp, introducing a comprehensive toolkit for speech processing including ASR, speaker diarization, and speech synthesis with support for multiple backends and APIs.
Full release notes