Wiki-Quellcode von Install OpenVoice v2 Locally - Instantly Clone Voice with AI

Version 2.1 von René Schmidt am 2025/03/10 17:57

author	version	line-number	content
		1	conda create -nopenvoice python=3.9
		2	conda activate openvoice
		3
		4
		5	git clone https:~/~/github.com/myshell-ai/openVoice.git
		6
		7
		8
		9
		10
		11	cd openVoice
		12	pip install -e .
		13	mkdir checkpoints_v2
		14	cd checkpoints_v2
		15	wget [[https:~~/~~/myshell-public-repo-host.s3.amazonaws.com/openvoice/checkpoints_v2_0417.zip>>https://myshell-public-repo-host.s3.amazonaws.com/openvoice/checkpoints_v2_0417.zip]]
		16
		17	unzip checkpoints_v2_0417.zip
		18
		19
		20	mv checkpoints_v2 ~~/openVoice
		21
		22
		23
		24
		25
		26
		27	pip install git+https:~/~/github.com/cocktailpeanut/ov2.git
		28
		29
		30
		31	cd ..
		32	pip install git+https:~/~/github.com/myshell-ai/MeloTTS.git
		33	python -m unidic download
		34	cd resources
		35	python3
		36
		37
		38
		39	import os
		40
		41	import torch
		42
		43	from openvoice import se_extractor
		44
		45	from openvoice.api import ToneColorConverter
		46
		47	ckpt_converter = 'checkpoints_v2/converter'
		48
		49	device = "cuda:0" if torch.cuda.is_available() else "cpu"
		50
		51	output_dir = 'outputs_v2'
		52
		53	tone_color_converter = ToneColorCOnverter(f'{ckpt_converter}/config.json', device=device)

Anwendungen