Parakeet.cpp – Parakeet ASR inference in pure C++ with Metal GPU acceleration

 (github.com)

31 points | by noahkay13 3 hours ago

4 comments

  • nullandvoid 7 minutes ago
    I've been using handy with parakeet on both Windows and mac, and have been very impressed.

    Hoe does this compare?

  • noahkay13 3 hours ago
    I built a C++ inference engine for NVIDIA's Parakeet speech recognition models using Axiom(https://github.com/Frikallo/axiom) my tensor library.

    What it does: - Runs 7 model families: offline transcription (CTC, RNNT, TDT, TDT-CTC), streaming (EOU, Nemotron), and speaker diarization (Sortformer) - Word-level timestamps - Streaming transcription from microphone input - Speaker diarization detecting up to 4 speakers

  • cakealert 27 minutes ago
    You should consider offering a python library.

    With models like these often you want to glue things together and manage multithreaded queues. And gluing in cpp is no fun.

    This assumes that your offering will perform better than onnxruntime on at least some metric such as memory.

  • ghostpepper 3 hours ago
    Off topic but if anyone is looking for a nice web-GUI frontend for a locally-hosted transcription engine, Scriberr is nice

    https://github.com/rishikanthc/Scriberr