mu-hashmi/personaplex-mlx
PersonaPlex on Apple Silicon: an MLX port of NVIDIA’s full-duplex speech-to-speech model with realtime local/web modes and offline WAV inference.
This tool helps you transform spoken audio into a different voice, making it sound like a full, natural conversation. You provide spoken audio and a text prompt defining the speaking style, and it outputs the same audio spoken in a new voice. Voice artists, content creators, or anyone needing to modify spoken dialogue for various applications would find this useful.
Use this if you need to quickly change the voice of spoken content while maintaining the original speech, especially for conversational AI or voice acting on an Apple Silicon Mac.
Not ideal if you need to remove background noise, perform echo cancellation, or use it on non-Apple Silicon hardware.
Stars
35
Forks
3
Language
Python
License
MIT
Category
Last pushed
Feb 18, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/mu-hashmi/personaplex-mlx"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
FluidInference/FluidAudio
Frontier CoreML audio models in your apps — text-to-speech, speech-to-text, voice activity...
k2-fsa/sherpa-ncnn
Real-time speech recognition and voice activity detection (VAD) using next-gen Kaldi with ncnn...
phuc-nt/my-translator
Real-time speech translation — macOS & Windows, free TTS, no server, your API keys only
pot-app/pot-desktop
🌈一个跨平台的划词翻译和OCR软件 | A cross-platform software for text translation and recognition.
Blaizzy/mlx-audio-swift
A modular Swift SDK for audio processing with MLX on Apple Silicon