skinny.

Latest / Kana & Mari's SoundRepos (English)

nyosegawa/speech.cpp | local speech synthesis and recognition | GitHub

speech.cpp is a C++ speech toolkit for running local text-to-speech, speech recognition, and voice activity detection models. It provides a command-line program, a C API and shared library, an OpenAI-compatible HTTP and Realtime transcription server, and a worker protocol; models are fetched by name and run through ggml. Related: nyosegawa/speech.cpp / local speech synthesis and recognition / GitHub

The skinny

The skinny isn't ready yet — notes appear once the transcript is processed.