Search Unity

  1. Welcome to the Unity Forums! Please take the time to read our Code of Conduct to familiarize yourself with the forum rules and how to post constructively.
  2. We have updated the language to the Editor Terms based on feedback from our employees and community. Learn more.
    Dismiss Notice

AudioStreamSpeechWhisper [offline speech recognition system]

Discussion in 'Assets and Asset Store' started by r618, Apr 15, 2023.

  1. r618

    r618

    Joined:
    Jan 19, 2009
    Posts:
    1,280
    : offline speech recognition, transcription, translation to English and language detection system based on originally OpenAI's Whisper, using an efficient whisper.cpp implementation running entirely locally on user device

    .either manual or automatic processing based on custom VAD (Voice Activity Detection) over audio stream (can be used in automatic 'open mic' fashion)

    an example running in macOS Editor:



    Please also see latest asset documentation
    Demo builds: Windows x64 | macOS | Linux (x64) | Android/ChromeOS
    Store page: Asset Store page
     
    Last edited: May 31, 2023
  2. r618

    r618

    Joined:
    Jan 19, 2009
    Posts:
    1,280
    Initial version - w/14 days new release discount - just went live -
     
  3. r618

    r618

    Joined:
    Jan 19, 2009
    Posts:
    1,280
    An update submitted, should be online hopefully shortly:

    mainly fixed models downloads, a VAD bugfix, and added CoreML support for Apple Silicon:

    V 1.4.7 092023 .250k
    - updated to (current) latest [1.4.7] Whisper.NET changes, also from now on following Whisper.NET versioning no.
    - updated model [HF] downloads & error handling
    - added models QuantizationType
    - updated iOS/macOS native libraries which now support CoreML
    - added automatic download of CoreML models
    - fixed VAD detection bug for open mic/continuos processing
    - see updated Docs for more about platform/macOS specific libraries