actableai/wav2vec2-ctc-khmer

from transformers import AutoModelForCTC, AutoProcessor, pipeline

if __name__ == "__main__":
    model_dir = "actableai/wav2vec2-ctc-khmer"

    processor = AutoProcessor.from_pretrained(model_dir, cache_dir="model-bin")
    model = AutoModelForCTC.from_pretrained(model_dir, cache_dir="model-bin")
    processor.tokenizer.pad_token = processor.tokenizer.bos_token
    model.config.pad_token_id = processor.tokenizer.pad_token_id

    asr_pipeline = pipeline(
        task="automatic-speech-recognition",
        model=model,
        tokenizer=processor.tokenizer,
        feature_extractor=processor.feature_extractor,
    )

    out = asr_pipeline("khmer_sample.wav")
    print(out)