← Back to Videos

A Fully Local Real-Time Voice Avatar

Loading video

audio.cpp combines LFM3-Audio 2B, SoulX-FlashHead and OWSM v4 into a local voice-to-avatar assistant with streaming speech, multilingual ASR and live animation.

Digital Human Speech Local Inference audio.cpp

Category: voice-interaction

Author: @0xShug0

Date: 2026-11-05T00:00:00

Duration: 41.717s

Source: https://x.com/0xShug0/status/2118366907367637296

LivLiv Player Ships a 34 KB Web Avatar

@yeko_L · 42 views · 2026-11-05

Digital HumanWeb Player

Falcon: Visual Target-Speaker Extraction in Noise

Falcon: Visual Target-Speaker Extraction in Noise

@aigclink · 123 views · 2026-10-31

Target Speaker ExtractionAudiovisual Learning

PersonaPlex Enables Real-Time Full-Duplex Voice Conversation

PersonaPlex Enables Real-Time Full-Duplex Voice Conversation

@liambraus · 157 views · 2026-10-22

Voice AIFull-Duplex Speech

ElevenLabs CLI v1: Voice Workflows in the Terminal

ElevenLabs CLI v1: Voice Workflows in the Terminal

@ElevenLabsDevs · 131 views · 2026-08-25

LiveKit Inference now runs cascaded, turn based and full duplex voice agents

LiveKit Inference now runs cascaded, turn based and full duplex voice agents

@livekit · 194 views · 2026-10-15

LiveKitVoice Agent

ElevenLabs introduces the Video to Music API

ElevenLabs introduces the Video to Music API

@ElevenLabsDevs · 170 views · 2026-10-15

ElevenLabsVideo to Music

As an Amazon Associate, we earn from qualifying purchases.