Whisper Language ID: Output all languages and their probabilities

[marchellodev]

Currently, the language ID based whisper outputs only one language with the top probability:

fn main() {
    let file_path = std::env::args().nth(1).expect("Missing file path argument");

    let (samples, sample_rate) = sherpa_rs::read_audio_file(&file_path).unwrap();
    assert_eq!(sample_rate, 16000, "The sample rate must be 16000.");

    let config = sherpa_rs::language_id::SpokenLanguageIdConfig {
        encoder: "sherpa-onnx-whisper-tiny/tiny-encoder.onnx".into(),
        decoder: "sherpa-onnx-whisper-tiny/tiny-decoder.onnx".into(),
        ..Default::default()
    };
    let mut extractor = sherpa_rs::language_id::SpokenLanguageId::new(config);

    let language = extractor.compute(samples, sample_rate).unwrap();
    println!("Spoken language: {}", language);
}

(from @sherpa-rs)

It would be great if sherpa could output the full result -- all languages and associated accuracies.

Faster-whisper, for example, can do this: https://github.com/SYSTRAN/faster-whisper/blob/c2a1da1bd94e002c38487c91c2f6b50a048000cf/faster_whisper/transcribe.py#L1764

[csukuangfj]

Would you like to contribute?