def main()

in ultravox/evaluation/string_metrics.py [0:0]

• 24 lines of code
• 3 McCabe index (conditional complexity)

def main():
    parser = argparse.ArgumentParser(
        description="Evaluate JSON files using WER and BLEU."
    )
    parser.add_argument("input_file", type=str, help="Path to the input JSON file.")
    parser.add_argument(
        "--metric",
        type=str,
        choices=["wer", "bleu"],
        required=True,
        help="Metric to compute.",
    )
    parser.add_argument(
        "--lang_id", type=str, default="en", help="Language ID (e.g., en, zh, ja)."
    )
    args = parser.parse_args()

    with open(args.input_file, "r", encoding="utf-8") as f:
        data = json.load(f)

    samples = [eval_types.Sample(**sample) for sample in data]

    if args.metric == "wer":
        result = wer(samples, {"lang_id": args.lang_id})
    else:
        result = bleu(samples, {"tokenize": args.lang_id})

    print(f"{args.metric.upper()} Score: {result.score}")