Speech Processing Python Packages

silero-vad

Silero VAD: pre-trained enterprise-grade Voice Activity Detector

whisper-timestamped

Multilingual Automatic Speech Recognition with word-level timestamps and confidence

torchscale

Foundation Architecture for (M)LLMs

indic-num2words

Python library for converting numbers to words for all Indian Languages.

pysptk

A python wrapper for Speech Signal Processing Toolkit (SPTK).

spafe

:sound: spafe: Simplified Python Audio Features Extraction

swift-f0

Fast and accurate fundamental frequency (F0) detector using convolutional neural networks

resemble-enhance

AI powered speech denoising and enhancement

nnmnkwii

Library to build speech synthesis systems designed for easy and fast prototyping.

voicefixer

General Speech Restoration

silero-vad-lite

Lightweight wrapper for Silero VAD using internal ONNX Runtime and with no python package dependencies

diarize

Speaker diarization for Python — "who spoke when?" CPU-only, no API keys, Apache 2.0. ~10.8% DER on VoxConverse, 8x faster than real-time.

stark-engine

S.T.A.R.K. - Speech And Text Algorithmic Recognition Kit

bournemouth-forced-aligner

Extract phoneme-level timestamps from speeh audio.

attenlabs-saa

Addressee detection for voice agents: device-directed speech detection that runs before STT, so background speech, side conversations, and the agent's own TTS echo never trigger it. No wake word, model-agnostic, drop-in for LiveKit, Pipecat, ElevenLabs, Twilio, and OpenAI. The layer your VAD and turn detection are missing.

saa-livekit-client

Addressee detection for voice agents: device-directed speech detection that runs before STT, so background speech, side conversations, and the agent's own TTS echo never trigger it. No wake word, model-agnostic, drop-in for LiveKit, Pipecat, ElevenLabs, Twilio, and OpenAI. The layer your VAD and turn detection are missing.

deepaudio-x

A python library to train Deep Neural Networks on various audio tasks using Self-Supervised backbones.

everyvoice

The EveryVoice TTS Toolkit - Text To Speech for your language

rutextnorm

Fast Russian Text normalization for TTS using only RegEx.

saa-pipecat-client

Addressee detection for voice agents: device-directed speech detection that runs before STT, so background speech, side conversations, and the agent's own TTS echo never trigger it. No wake word, model-agnostic, drop-in for LiveKit, Pipecat, ElevenLabs, Twilio, and OpenAI. The layer your VAD and turn detection are missing.

russian-tts-normalization

Fast Russian Text normalization for TTS using only RegEx.

scoreq

SCOREQ: Speech COntrastive REgression for Quality Assessment (NeurIPS 2024)

vak

A neural network framework for researchers studying acoustic communication

voicetut-tts

VoiceTut-TTS is an Egyptian-Arabic text-to-speech system fine-tuned from OmniVoice on ~380 hours of Egyptian podcast speech. It produces natural Egyptian speech with seamless Arabic ↔ English code-switching, ships 15 built-in studio voices, supports zero-shot voice cloning