webrtcvad-wheels

What it is and what it does

This package wraps Google's WebRTC Voice Activity Detector, a fast and accurate classifier that determines whether a short audio segment contains speech or silence. It accepts 16-bit mono PCM audio at standard sample rates and returns a boolean result, with tunable aggressiveness (0–3) to control sensitivity to non-speech sounds. The package is distributed with pre-compiled binary wheels, eliminating the need to build the underlying C extension on Windows, macOS, and Linux across multiple architectures and CPU types.

The VAD is commonly used as a preprocessing step in speech recognition pipelines, telephony systems, and audio analysis workflows where you need to filter out silence or identify voiced regions before further processing. With no runtime dependencies and a simple API, it integrates easily into Python audio applications.

Use it for:

• Filter silence from audio recordings before sending to a speech-to-text service to reduce processing cost and latency.
• Segment a continuous audio stream into voiced and unvoiced regions for telephony or voice call analysis.
• Preprocess microphone input in real-time speech recognition to skip processing during silence.
• Detect speech activity in surveillance or meeting recordings to identify when participants are speaking.
• Build a voice activity detector for audio quality assessment or speaker diarization pipelines.

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Detects voiced versus unvoiced segments in audio by wrapping Google's WebRTC Voice Activity Detector with pre-built binary wheels for Windows, macOS, and Linux.

Yes. The package is stable (Production/Stable status), has no known vulnerabilities, zero runtime dependencies, and ships with pre-built wheels that eliminate compilation friction on major platforms. The aging maintenance status is acceptable for a mature, focused tool with a narrow scope. Install it if you need reliable voice activity detection in Python.

Install

webrtcvad-wheels on PyPI

pip install webrtcvad-wheels

uv add webrtcvad-wheels

poetry add webrtcvad-wheels

Installing webrtcvad-wheels

Before you install

Medium install friction due to compiled C extension, but mitigated by pre-built wheels covering Windows, macOS, and Linux across multiple architectures. Repository is aging (708 days since last release) but remains actively maintained with recent commits.

License in practice

MIT license permits commercial and private use with minimal restrictions; you must include the license text in distributions.

Quickstart

pip install webrtcvad-wheels

import webrtcvad
vad = webrtcvad.Vad(1)  # aggressiveness 0-3
frame = b'\x00\x00' * int(16000 * 10 / 1000)  # 10ms at 16kHz
is_speech = vad.is_speech(frame, 16000)

Audio must be 16-bit mono PCM at 8000, 16000, 32000, or 48000 Hz; frames must be exactly 10, 20, or 30 ms in duration.

Verify before relying

• Whether the package supports Python versions below 3.6 despite classifiers listing 3.6+
• Exact memory leak fixes and performance improvements claimed in version history beyond what the fact sheet documents
• Specific audio frame duration requirements and sample rate constraints beyond the documented 10, 20, or 30 ms frames

Package facts

Evidence: webrtcvad_wheels-2.0.14-cp310-cp310-macosx_10_9_x86_64.whl; webrtcvad_wheels-2.0.14-cp310-cp310-macosx_11_0_arm64.whl; webrtcvad_wheels-2.0.14-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl; webrtcvad_wheels-2.0.14-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl; webrtcvad_wheels-2.0.14-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl; webrtcvad_wheels-2.0.14-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl; webrtcvad_wheels-2.0.14-cp310-cp310-musllinux_1_2_aarch64.whl; webrtcvad_wheels-2.0.14-cp310-cp310-musllinux_1_2_i686.whl; webrtcvad_wheels-2.0.14-cp310-cp310-musllinux_1_2_ppc64le.whl; webrtcvad_wheels-2.0.14-cp310-cp310-musllinux_1_2_x86_64.whl; webrtcvad_wheels-2.0.14-cp310-cp310-win32.whl; webrtcvad_wheels-2.0.14-cp310-cp310-win_amd64.whl; webrtcvad_wheels-2.0.14-cp311-cp311-macosx_10_9_x86_64.whl; webrtcvad_wheels-2.0.14-cp311-cp311-macosx_11_0_arm64.whl; webrtcvad_wheels-2.0.14-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl; webrtcvad_wheels-2.0.14-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl; webrtcvad_wheels-2.0.14-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl; webrtcvad_wheels-2.0.14-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl; webrtcvad_wheels-2.0.14-cp311-cp311-musllinux_1_2_aarch64.whl; webrtcvad_wheels-2.0.14-cp311-cp311-musllinux_1_2_i686.whl

Keywords: speechrecognition, asr, voiceactivitydetection, vad, webrtc

Tags

More Artificial Intelligence packages