Punctuation failed with CoreML

[thewh1teagle]

It failed with CoreML on macOS but works with CPU.

Log:

cargo run --example punctuate
   Compiling sherpa-rs v0.5.1-beta.0 (/Volumes/Internal/sherpa-rs)
    Finished `dev` profile [unoptimized + debuginfo] target(s) in 0.30s
     Running `target/debug/examples/punctuate`
/Users/runner/work/sherpa-onnx/sherpa-onnx/sherpa-onnx/c-api/c-api.cc:SherpaOnnxCreateOfflinePunctuation:1589 OfflinePunctuationConfig(model=OfflinePunctuationModelConfig(ct_transformer="./sherpa-onnx-punct-ct-transformer-zh-en-vocab272727-2024-04-12/model.onnx", num_threads=1, debug=True, provider="coreml"))

/Users/runner/work/sherpa-onnx/sherpa-onnx/sherpa-onnx/csrc/offline-ct-transformer-model.cc:Init:117 
vocab_size: 272727
num_punctuations: 6
punctuations: <unk> _ ， 。 ？ 、 


--------------------
2024-10-20 02:46:26.496 punctuate[784:1415639] 2024-10-20 02:46:26.496798 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running CoreML_15887252831431197149_7 node. Name:'CoreMLExecutionProvider_CoreML_15887252831431197149_7_7' Status Message: Error executing model: Unable to compute the prediction using a neural network model. It can be an invalid input data or broken/unsupported model (error code: -1).
fatal runtime error: Rust cannot catch foreign exceptions
[1]    784 abort      cargo run --example punctuate

Code:

/*
We assume you have pre-downloaded the model files for testing
from https://github.com/k2-fsa/sherpa-onnx/releases/tag/punctuation-models

wget https://github.com/k2-fsa/sherpa-onnx/releases/download/punctuation-models/sherpa-onnx-punct-ct-transformer-zh-en-vocab272727-2024-04-12.tar.bz2
tar xvf sherpa-onnx-punct-ct-transformer-zh-en-vocab272727-2024-04-12.tar.bz2
rm sherpa-onnx-punct-ct-transformer-zh-en-vocab272727-2024-04-12.tar.bz2

cargo run --example punctuate
*/

fn main() {
    let model = "./sherpa-onnx-punct-ct-transformer-zh-en-vocab272727-2024-04-12/model.onnx";
    let sentences = [
        "这是一个测试你好吗How are you我很好thank you are you ok谢谢你", 
        "我们都是木头人不会说话不会动", 
        "The African blogosphere is rapidly expanding bringing more voices online in the form of commentaries opinions analyses rants and poetry"
    ];

    let config = sherpa_rs::punctuate::PunctuationConfig {
        model: model.into(),
        debug: true,
        ..Default::default()
    };
    let mut punctuate = sherpa_rs::punctuate::Punctuation::new(config).unwrap();

    println!("--------------------");
    for sentence in sentences {
        let punctuated = punctuate.add_punctuation(&sentence);
        println!("Input text: {}", sentence);
        println!("Output text: {}", punctuated);
        println!("--------------------");
    }
}

[csukuangfj]

I suggest that you use cpu at present.
Not every model supports coreml.