export

游雁

2023-03-31 4ba1011b42e041ee1d71448eefd7ef2e7bd61bb6

 funasr/tasks/asr.py

@@ -125,7 +125,7 @@
        bicif_paraformer=BiCifParaformer,
        contextual_paraformer=ContextualParaformer,
        mfcca=MFCCA,
        timestamp_predictor=TimestampPredictor,
        timestamp_prediction=TimestampPredictor,
    ),
    type_check=AbsESPnetModel,
    default="asr",
@@ -411,6 +411,12 @@
            type=str,
            default="13_15",
            help="The range of noise decibel level.",
        )
        parser.add_argument(
            "--batch_interval",
            type=int,
            default=10000,
            help="The batch interval for saving model.",
        )

        for class_choices in cls.class_choices_list:
@@ -826,7 +832,7 @@
            if "model.ckpt-" in model_name or ".bin" in model_name:
                model_name_pth = os.path.join(model_dir, model_name.replace('.bin',
                                                                            '.pb')) if ".bin" in model_name else os.path.join(
                    model_dir, "{}.pth".format(model_name))
                    model_dir, "{}.pb".format(model_name))
                if os.path.exists(model_name_pth):
                    logging.info("model_file is load from pth: {}".format(model_name_pth))
                    model_dict = torch.load(model_name_pth, map_location=device)
@@ -1073,7 +1079,7 @@
            if "model.ckpt-" in model_name or ".bin" in model_name:
                model_name_pth = os.path.join(model_dir, model_name.replace('.bin',
                                                                            '.pb')) if ".bin" in model_name else os.path.join(
                    model_dir, "{}.pth".format(model_name))
                    model_dir, "{}.pb".format(model_name))
                if os.path.exists(model_name_pth):
                    logging.info("model_file is load from pth: {}".format(model_name_pth))
                    model_dict = torch.load(model_name_pth, map_location=device)
@@ -1278,8 +1284,6 @@
            token_list = list(args.token_list)
        else:
            raise RuntimeError("token_list must be str or list")
        vocab_size = len(token_list)
        logging.info(f"Vocabulary size: {vocab_size}")

        # 1. frontend
        if args.input_size is None:
@@ -1316,6 +1320,7 @@
            frontend=frontend,
            encoder=encoder,
            predictor=predictor,
            token_list=token_list,
            **args.model_conf,
        )

@@ -1332,12 +1337,3 @@
    ) -> Tuple[str, ...]:
        retval = ("speech", "text")
        return retval


class ASRTaskAligner(ASRTaskParaformer):
    @classmethod
    def required_data_names(
            cls, train: bool = True, inference: bool = False
    ) -> Tuple[str, ...]:
        retval = ("speech", "text")
        return retval