audio-cpp

/

audio.cpp-gguf

Like 200

@WhatABigWorld I’ve reached out to the model author to see if he could help with the integration. I can’t really tell how good the model’s output quality is for Arabic. This has happened before with other languages as well. I thought they were working correctly based on the ASR results, but it turned out they weren’t...

Hi @audio-cpp I'm native arabic speaker and i tested the model and it's Egyptian dialect not modern standard arabic, i also tested it with pure arabic text and it produces output with highly Egyptian accent influence, not worth adding

VoiceTut-TTS and QwenCleo-ASR were added by their author as variants. They’re fine-tuned versions of models we already support, so adding them only required a few lines of changes. I think it’s good to offer users more choices.