add sond model

志浩

2023-02-10 f6a1cdaf3488c9ec572e1f753b50cb58a0f8fd79

 funasr/bin/sv_inference.py

@@ -1,4 +1,7 @@
#!/usr/bin/env python3
# Copyright FunASR (https://github.com/alibaba-damo-academy/FunASR). All Rights Reserved.
#  MIT License  (https://opensource.org/licenses/MIT)

import argparse
import logging
import os
@@ -26,7 +29,7 @@
from funasr.utils.types import str2bool
from funasr.utils.types import str2triple_str
from funasr.utils.types import str_or_none

from funasr.utils.misc import statistic_model_parameters

class Speech2Xvector:
    """Speech2Xvector class
@@ -59,6 +62,7 @@
            device=device
        )
        logging.info("sv_model: {}".format(sv_model))
        logging.info("model parameter number: {}".format(statistic_model_parameters(sv_model)))
        logging.info("sv_train_args: {}".format(sv_train_args))
        sv_model.to(dtype=getattr(torch, dtype)).eval()

@@ -156,17 +160,17 @@


def inference_modelscope(
        output_dir: Optional[str],
        batch_size: int,
        dtype: str,
        ngpu: int,
        seed: int,
        num_workers: int,
        log_level: Union[int, str],
        key_file: Optional[str],
        sv_train_config: Optional[str],
        sv_model_file: Optional[str],
        model_tag: Optional[str],
        output_dir: Optional[str] = None,
        batch_size: int = 1,
        dtype: str = "float32",
        ngpu: int = 1,
        seed: int = 0,
        num_workers: int = 0,
        log_level: Union[int, str] = "INFO",
        key_file: Optional[str] = None,
        sv_train_config: Optional[str] = "sv.yaml",
        sv_model_file: Optional[str] =  "sv.pth",
        model_tag: Optional[str] = None,
        allow_variable_data_keys: bool = True,
        streaming: bool = False,
        embedding_node: str = "resnet1_dense",
@@ -214,7 +218,6 @@
            data_path_and_name_and_type: Sequence[Tuple[str, str, str]] = None,
            raw_inputs: Union[np.ndarray, torch.Tensor] = None,
            output_dir_v2: Optional[str] = None,
            fs: dict = None,
            param_dict: Optional[dict] = None,
    ):
        logging.info("param_dict: {}".format(param_dict))