Generating Kokoro TTS audio with coreml provider

[l3utterfly]

I got this error generating TTS audio with provider = "coreml"

2025-02-04 22:24:47.002908 [I:onnxruntime:, model_builder.cc:926 SaveModel] Writing CoreML Model to /private/var/mobile/Containers/Data/Application/9F1EE4F3-7EC8-4572-A177-A6237E1E8E2F/tmp/onnxruntime-9F750DCD-7DEF-46C5-991E-2E7CC3663DA4-29393-00001A50FC386A09.model.mlmodel
libc++abi: terminating due to uncaught exception of type Ort::Exception: model_builder.cc:760 RegisterModelInputOutput Unable to get shape for output: /Squeeze_output_0

This is done on an iPhone 15 Pro
Using provided Kokoro onnx model in this repo

[csukuangfj]

Please use provider=cpu first.

[l3utterfly]

Is coreml provider planned?

[csukuangfj]

That is an issue of onnxruntime.
Please first try if you can use onnxruntime python package with coreml with this model.

[l3utterfly]

Alright, thank you very much!

[ahadjawaid]

Actually, you can run the newest kokoro model, v1, using coreml. But, I haven't really seen any speed ups with it.

[l3utterfly]

Actually, you can run the newest kokoro model, v1, using coreml. But, I haven't really seen any speed ups with it.

Where can I download this?

[ahadjawaid]

Here is the download link.