OpenVoice
Instant voice cloning from a short sample, with tone and accent carried across languages.
Voiceovers and multilingual dubbing in a client's own voice.
MIT, but last pushed April 2025. Cloning a voice needs the speaker's written consent.
MIT (LICENSE: 'Copyright 2024 MyShell.ai' + MIT text; README: 'OpenVoice V1 and V2 are MIT Licensed. Free for both commercial and research use').
Yes: README states both V1 and V2 (code and models) are MIT and free for commercial use since April 2024.
Looks unmaintained: last push 2025-04-19, and V2 dates from April 2024. Voice cloning from a short reference clip: legal risk if a real person's voice is cloned without documented consent (impersonation, publicity/biometric laws); neither the README nor the research page gives any consent or misuse policy. GPU/VRAM requirement: not found in the README or research page. Not archived.
Course creators, advertisers, podcasters
Voiceovers and multilingual dubbing in a client's own voice.
none found for OpenVoice itself (the research page has no pricing; README says it powers voice cloning on the vendor's platform myshell.ai, whose pricing was not checked).
None named