[HN Gopher] Amphion: An open-source audio, music, and speech gen...
       ___________________________________________________________________
        
       Amphion: An open-source audio, music, and speech generation toolkit
        
       Author : lapnect
       Score  : 77 points
       Date   : 2024-10-27 14:49 UTC (8 hours ago)
        
 (HTM) web link (github.com)
 (TXT) w3m dump (github.com)
        
       | 4b11b4 wrote:
       | this repo is just another point in the bucket of open source "AI"
       | (and all the tools around it)
       | 
       | is the only answer
        
       | thot_experiment wrote:
       | I'll definitely be checking this out but if anyone has
       | recommendations on a TTS system that generates good output and
       | has well documented voice training tools I'm interested in this
       | field and would like to dive in again. At one point I was using
       | https://git.ecker.tech/mrq/ai-voice-cloning/wiki/Installatio...
       | and llama to generate Obama speeches commemorating our DotA
       | games, but the whole thing was very brittle, slow to generate and
       | training voices was also slow and hit or miss.
       | 
       | This was a couple years ago but every time I look it doesn't seem
       | like the field has any equivalent to ollama/ooba or
       | auto1111/comfy, though perhaps this will be it. I'm all ears wrt
       | recommendations! (also interested in neural RVC, would be
       | extremely useful for wreaking havoc on my discord :3)
        
       ___________________________________________________________________
       (page generated 2024-10-27 23:01 UTC)