RealtimeSTT

RealtimeSTT is a Python speech-to-text library that provides voice activity detection, fast transcription, and wake words.

RealtimeSTT is a Python speech-to-text library designed for assistants, dictation tools, and streaming applications. It turns speech into text using various engines such as faster-whisper, sherpa-onnx, and Kroko-ONNX. The library supports microphone input as well as external audio streams like files and WebSockets. It includes features like voice activity detection, wake word activation, event callbacks, and a production FastAPI server. Developers use it to build interactive voice applications with minimal code.

RealtimeSTT at a glance

Best for: Software engineer
Runs on: Linux, macOS, Windows, self-hosted, open source, API
Pricing: Free
Released: August 2023
Category: AI audio tool
Website: github.com/KoljaB/RealtimeSTT

Key features

Voice activity detection with WebRTC and Silero VAD
Final and realtime transcription with selectable engines
Wake word activation via Porcupine or OpenWakeWord
Direct microphone input or external audio stream feeding
Event callbacks for recording and transcription states
Packaged FastAPI server with HTTP and WebSocket endpoints

Use it for

Build voice assistants and dictation tools
Stream live audio for speech transcription
Process external audio files or WebSocket streams
Deploy remote speech-to-text servers

Alternatives to RealtimeSTT

RealtimeTTS — Provides low-latency speech synthesis with a broad engine matrix and multiple backends.
Whisper Timestamped — Generates accurate word-level timestamps without extra neural networks.
WhisperLive — It provides real-time speech-to-text using multiple backends like Faster Whisper and TensorRT.
WeNet — Achieves state-of-the-art results on public speech datasets.
AssemblyAI — Industry-leading speech models deliver high accuracy and low latency for voice applications.
faster-whisper — Transcribes audio faster with lower memory usage than standard implementations.
Speech-to-Speech — Fully modular, swappable components with local or hosted execution options.
ESPnet — Provides reproducible recipes and hundreds of pretrained models for various speech tasks.
Sys2txt — Transcribes system audio locally using flexible Whisper engines.
talkflow — Transcribes speech entirely on your device with no word limits or cloud accounts
Esp32 Pie ASR — Faster than real time streaming speech recognition on a microcontroller alone
FunASR — It supports offline, streaming, and edge deployment with multiple models.

All RealtimeSTT alternatives