<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Ml-Inference on Noureddine RAMDI</title><link>https://ramdi.fr/tags/ml-inference/</link><description>Recent content in Ml-Inference on Noureddine RAMDI</description><generator>Hugo</generator><language>en</language><lastBuildDate>Sat, 23 May 2026 20:41:27 +0000</lastBuildDate><atom:link href="https://ramdi.fr/tags/ml-inference/index.xml" rel="self" type="application/rss+xml"/><item><title>QwenVoice: offline Apple Silicon text-to-speech with XPC isolation and model quantization tradeoffs</title><link>https://ramdi.fr/github-stars/qwenvoice-offline-apple-silicon-text-to-speech-with-xpc-isolation-and-model-quantization-tradeoffs/</link><pubDate>Mon, 04 May 2026 10:23:02 +0000</pubDate><guid>https://ramdi.fr/github-stars/qwenvoice-offline-apple-silicon-text-to-speech-with-xpc-isolation-and-model-quantization-tradeoffs/</guid><description>QwenVoice runs Qwen3-TTS 1.7B offline on Apple Silicon using MLX with XPC isolation and supports voice cloning. It balances 8-bit quality and 4-bit speed models in a native macOS/iOS app.</description></item></channel></rss>