realtimestt
A fast Voice Activity Detection and Transcription System
Decision gist · record as of 2026-08-14
Yes. RealtimeSTT is actively maintained, has no known vulnerabilities, and offers a clean API for real-time speech-to-text with flexible backend selection. The permissive MIT license and low install friction (pure wheel, optional extras) make it low-risk. The main gotchas are the system PortAudio dependency and Python 3.11+ requirement; if your environment meets those, it's a solid choice for local speech recognition without cloud APIs.AI-flagged interpretation of the facts on this page — verify before relying
Before you install
- Requires Python 3.11+, system PortAudio headers (apt-get install portaudio19-dev on Linux, brew install portaudio on macOS), and torch/torchaudio installed via the extras.
- Low install friction; pure Python wheel.
- Requires system PortAudio headers (apt-get on Linux, brew on macOS) and Python 3.11+.
License · maintenance · safety
MIT (permissive) — MIT license is permissive; you can use, modify, and distribute this package freely in commercial or private projects with minimal restrictions.
last release 2026-05-31 (75 days) · last repo commit 2026-06-12 · 10,055 stars
0 known vulnerabilities (OSV.dev, 2026-08-14) · 111,392 downloads/mo, #12,417 on PyPI
Alternatives
Verify before relying
pip install "realtimestt[faster-whisper]"
from RealtimeSTT import AudioToTextRecorder
if __name__ == "__main__":
with AudioToTextRecorder() as recorder:
print("Speak now")
print(recorder.text())- Whether the package's multiprocessing implementation works reliably across all platforms and Python patch versions.
- Real-world latency and accuracy benchmarks compared to other local speech-to-text solutions.
- Stability and completeness of less-common backend integrations (Kroko, Omnilingual, Parakeet).
What it is and what it does
RealtimeSTT is a Python library that captures audio from a microphone or external stream, detects when speech is happening using voice activity detection (VAD), and transcribes it to text. It's designed for applications like voice assistants, dictation tools, and streaming servers that need to turn speech into text with minimal code. The library supports multiple transcription backends (faster-whisper, OpenAI Whisper, Silero, and others) via optional extras, so you install only what you need. It includes WebRTC VAD by default and can optionally use Silero VAD for better accuracy, plus optional wake-word detection through Porcupine or OpenWakeWord.
The core package is lightweight, but it depends on PyAudio, torch, torchaudio, scipy, websockets, and other audio libraries. Installation requires system-level PortAudio headers on Linux and macOS. Python 3.11 or newer is required. The package uses multiprocessing for model work, which means you must guard your main code with `if __name__ == "__main__":` on Windows. You can feed audio from files, streams, or websockets by setting `use_microphone=False` and calling `feed_audio()` with PCM chunks.
Use it for
- Build a voice assistant that listens for a wake word, then transcribes commands in real-time with minimal latency.
- Create a dictation tool that continuously records speech and processes transcripts asynchronously via callbacks.
- Stream audio from a websocket or external source and transcribe it without touching the local microphone.
- Prototype a speech-to-text pipeline where you can swap transcription engines (faster-whisper, Kroko, Silero) without changing application code.
- Detect ambient noise levels and voice activity to trigger recording or filtering in audio processing pipelines.
Worth the install?
AI-flagged interpretation of the facts on this page. Verify before relying on it.
Yes.
RealtimeSTT is actively maintained, has no known vulnerabilities, and offers a clean API for real-time speech-to-text with flexible backend selection. The permissive MIT license and low install friction (pure wheel, optional extras) make it low-risk. The main gotchas are the system PortAudio dependency and Python 3.11+ requirement; if your environment meets those, it's a solid choice for local speech recognition without cloud APIs.
Install
realtimestt on PyPI
Before you install
Low install friction; pure Python wheel. Requires system PortAudio headers (apt-get on Linux, brew on macOS) and Python 3.11+. Nine runtime dependencies including torch, torchaudio, and PyAudio; optional extras let you choose transcription backends. Actively maintained with recent commits and 10055 GitHub stars.
Requires Python 3.11+, system PortAudio headers (apt-get install portaudio19-dev on Linux, brew install portaudio on macOS), and torch/torchaudio installed via the extras.
License in practice
MIT license is permissive; you can use, modify, and distribute this package freely in commercial or private projects with minimal restrictions.
Quickstart
pip install "realtimestt[faster-whisper]"
from RealtimeSTT import AudioToTextRecorder
if __name__ == "__main__":
with AudioToTextRecorder() as recorder:
print("Speak now")
print(recorder.text())Verify before relying
- Whether the package's multiprocessing implementation works reliably across all platforms and Python patch versions.
- Real-world latency and accuracy benchmarks compared to other local speech-to-text solutions.
- Stability and completeness of less-common backend integrations (Kroko, Omnilingual, Parakeet).
Package facts
| License | MIT permissive |
| Python support | Supports the current Python release >=3.11 |
| Install friction | Low. Pure-Python wheel |
| Runtime dependencies | 9 packagesPyAudiowebrtcvad-wheelshalotorchtorchaudioscipywebsocketswebsocket-clientsoundfile |
| Maintenance | Actively maintained 75 days since the last release |
| Last repo commit | |
| First released | |
| Downloads | 111,392 / month, #12,417 on PyPI 30-day window, as of 2026-08-14 |
| Known vulnerabilities | None known OSV.dev, checked 2026-08-14 |
Evidence: realtimestt-1.0.2-py3-none-any.whl
Tags
Let your AI agent find packages like this
Example. Real query, live index.
You found this page by searching. An agent finds it by wishing: SkillFed indexes 14,416 PyPI packages by what they can do, searchable in plain language.
wish › “real-time transcription engine”
- realtimesttRealtimeSTT converts speech from microphone or external audio streams…
- pocketsphinxPocketSphinx provides Python bindings for Carnegie Mellon…
- speechmatics-rtAsync Python client for real-time speech-to-text transcription via…
Give your agent the search over MCP, or paste the wish link into any chat.
More Artificial Intelligence packages
LiteLLM provides a unified Python interface to call 100+ LLM providers (OpenAI, Anthropic, Gemini, Bedrock, Azure, and others) using OpenAI-compatible API format, available as both a Python SDK and a self-hosted AI Gateway proxy server.
Install it if you need to work with multiple LLM providers or want to centralize LLM routing in your organization.
Client library and CLI tool for downloading, uploading, and managing models, datasets, and repositories on the Hugging Face Hub platform.
Install it if you work with Hugging Face Hub models or datasets.
LangChain provides a framework for building agents and LLM-powered applications by composing language models, tools, and memory through a unified API that abstracts over multiple model providers.
hf-xet provides chunk-based deduplication and efficient file transfer for the Hugging Face Hub, enabling faster uploads and downloads of large files with local disk caching.
Tokenizers converts raw text into token sequences for NLP models, with support for training custom vocabularies and using pre-built tokenizers (BPE, WordPiece) optimized for speed via Rust.
Transformers provides a unified framework for loading, fine-tuning, and running state-of-the-art pretrained models across text, vision, audio, video, and multimodal tasks using PyTorch, JAX, or TensorFlow.
Install it if you need to run or train any transformer-based model for NLP, vision, audio, or multimodal tasks.
See also faster-whisper · openwakeword · pvporcupine · wyoming · whisper-timestamped · whisperx · silero-vad · SpeechRecognition · pymicro-vad · webrtcvad