jamsch/expo-speech-recognition

Speech Recognition for React Native Expo projects

What it solves

expo-speech-recognition provides a unified API for implementing speech-to-text functionality in React Native projects. It solves the problem of fragmented native speech recognition APIs across iOS, Android, and the Web, allowing developers to write a single codebase that works across all three platforms.

How it works

The library acts as a wrapper around the native speech recognition services of each platform:

  • iOS: Uses SFSpeechRecognizer.
  • Android: Uses SpeechRecognizer.
  • Web: Uses the SpeechRecognition API.

It provides a set of hooks (like useSpeechRecognitionEvent) and a direct module API (ExpoSpeechRecognitionModule) to start, stop, and manage speech recognition sessions. It handles permissions, event-driven results (interim and final), and platform-specific configurations such as on-device recognition and audio session categories.

Who it’s for

React Native and Expo developers who need to integrate voice input or transcription services into their mobile and web applications without writing platform-specific native code.

Highlights

  • Cross-platform support: Unified interface for iOS, Android, and Web.
  • Flexible transcription: Supports both real-time (interim) results and final transcriptions.
  • Audio persistence: Ability to save audio recordings to local files on Android 13+ and iOS.
  • File-based transcription: Supports transcribing existing audio files from a local URI.
  • Advanced controls: Includes volume metering, language detection (Android 14+), and the ability to to force on-device recognition to prevent network audio transmission.
  • Customization: Allows providing contextual strings to improve recognition accuracy for specific app-related terms.

Related

  • Project
    deeeed/audiolab

    An audio processing monorepo for React Native and Expo that provides on-device speech-to-text, text-to-speech, and audio analysis across iOS, Android, and web.

  • Project
    mybigday/whisper.rn

    React Native bindings for whisper.cpp, enabling on-device speech-to-text using Whisper and NVIDIA Parakeet models for iOS and Android.

  • Project
    codeforequity-at/botium-speech-processing

    A unified API for open-source and cloud-based Speech-To-Text and Text-To-Speech services, simplifying audio processing for chatbots and voice applications.

  • Project
    algolia/voice-overlay-ios

    An iOS library that provides a polished voice-to-text overlay UI, handling permissions and speech recognition using Apple's native SFSpeechRecognizer.

  • Project
    Uberi/speech_recognition

    A Python library that provides a unified interface for speech recognition, supporting multiple online and offline engines like OpenAI Whisper, Google Speech, and Vosk.