Article suggests text-to-speech model output format is an image

[chowieuk]

Existing documentation URL(s)

https://developers.cloudflare.com/workers-ai/models/aura-1/

Output

The binding returns a ReadableStream with the image in JPEG or PNG format (check the model's output schema).

What changes are you suggesting?

I haven't worked with this worker / model, but I'm assuming the output is audio.

Additional information

No response

[jonom]

I spent quite a while trying to figure this out but couldn't get anywhere. Eventually got @cf/myshell-ai/melotts working instead - the 'ah-ha' moment was discovering that the returned audio is in wav format despite the docs saying it is mp3!