News for hackers. Stuff that matters.

A comment on OpenAI Audio Models. Parent comment

Re:OpenAI Audio Models by whimsicalism on Thursday March 20, 2025 @04:05PM (#43428244)
Sesame is free and pretty good and you can run it yourself.
[ Parent | on HN ]
Re:OpenAI Audio Models by kuprel on Thursday March 20, 2025 @04:27PM (#43428506)
They released a crippled model: https://github.com/SesameAILabs/csm/issues/63
[ Parent | on HN ]
Re:OpenAI Audio Models by hnhn34 on Thursday March 20, 2025 @04:48PM (#43428764)
The good news is Orpheus-3B just made Sesame essentially obsolete.
[ Parent | on HN ]
Re:OpenAI Audio Models by Foreignborn on Thursday March 20, 2025 @05:50PM (#43429387)
thanks for this, it sounds pretty good.

link for anyone else: https://canopylabs.ai/model-releases

[ Parent | on HN ]
Re:OpenAI Audio Models by sandspar on Thursday March 20, 2025 @06:24PM (#43429727)
These voices are all annoying, though. The thing about Sesame's Miles is that he's cool.
[ Parent | on HN ]
Re:OpenAI Audio Models by satvikpendem on Friday March 21, 2025 @02:50PM (#43439565)
You can clone any voice you want with Orpheus.
[ Parent | on HN ]
Re:OpenAI Audio Models by whimsicalism on Friday March 21, 2025 @01:56PM (#43438968)
admittedly i only tested it for a few queries, but it worked fine for me except for lack of consistency in voice
[ Parent | on HN ]