mybigday/whisper.rn

React Native binding of whisper.cpp.

What it solves

whisper.rn provides React Native bindings for high-performance, on-device automatic speech recognition (ASR). It allows developers to integrate OpenAI's Whisper and NVIDIA Parakeet models into mobile applications without relying on cloud-based transcription services.

How it works

The library acts as a bridge to whisper.cpp, a C++ implementation of Whisper. It supports loading GGML model binaries (including quantized versions) and provides an asynchronous API for transcribing audio files or raw PCM data. It also includes support for Voice Activity Detection (VAD) via the Silero VAD model to identify speech segments in audio.

Who it’s for

Mobile developers using React Native who need to implement offline, private, and low-latency transcription or real-time speech-to-text capabilities in their iOS and Android apps.

Highlights

  • Multi-Model Support: Supports both OpenAI's Whisper and NVIDIA Parakeet TDT models.
  • Real-time Transcription: Includes a RealtimeTranscriber for streaming audio transcription with integrated VAD and auto-slicing.
  • Hardware Acceleration: Supports Core ML for accelerated inference on iOS devices.
  • VAD Integration: Built-in support for Silero VAD to detect speech segments from files, URLs, or raw data.
  • Flexible Asset Management: Models and audio files can be loaded from the local filesystem or bundled as app assets.

Related

  • Project
    sandrohanea/whisper.net

    Dotnet bindings for OpenAI Whisper via whisper.cpp, enabling local speech-to-text and translation in .NET applications with extensive hardware acceleration support.

  • Project
    ggml-org/whisper.cpp

    A high-performance C/C++ port of OpenAI's Whisper model for efficient, offline, on-device automatic speech recognition across multiple platforms.

  • Project
    ahmetoner/whisper-asr-webservice

    A general-purpose speech recognition toolkit that wraps Whisper models into a REST API for easy deployment and integration.

  • Project
    jamsch/expo-speech-recognition

    A cross-platform React Native library that provides a unified API for speech-to-text functionality across iOS, Android, and the Web.

  • Project
    collabora/WhisperLive

    WhisperLive is an open‑source, near‑real‑time transcription server built around OpenAI’s Whisper model. It supports multiple inference back‑ends (faster‑whisper, NVIDIA TensorRT, Intel OpenVINO, AMD ROCm), offers word‑level timestamps, hot‑word boosting, speaker diarization and optional translation, and provides both simple command‑line clients and a Python streaming API. Docker images are provided for GPU and CPU deployments, and browser/iOS extensions let you transcribe audio directly from a web page or mobile device.