support asr_inference_paraformer_vad_punc

funasr/bin/asr_inference_paraformer_vad_punc.py

@@ -144,7 +144,7 @@ class Speech2Text:
         for scorer in scorers.values():
             if isinstance(scorer, torch.nn.Module):
                 scorer.to(device=device, dtype=getattr(torch, dtype)).eval()
-        
+
         logging.info(f"Decoding device={device}, dtype={dtype}")
 
         # 5. [Optional] Build Text converter: e.g. bpe-sym -> Text
@@ -184,12 +184,11 @@ class Speech2Text:
         self.encoder_downsampling_factor = 1
         if asr_train_args.encoder_conf["input_layer"] == "conv2d":
             self.encoder_downsampling_factor = 4
-        
-            
 
     @torch.no_grad()
     def __call__(
-            self, speech: Union[torch.Tensor, np.ndarray], speech_lengths: Union[torch.Tensor, np.ndarray] = None, begin_time: int = 0, end_time: int = None, 
+            self, speech: Union[torch.Tensor, np.ndarray], speech_lengths: Union[torch.Tensor, np.ndarray] = None,
+            begin_time: int = 0, end_time: int = None,
     ):
         """Inference
 
@@ -215,7 +214,7 @@ class Speech2Text:
         else:
             feats = speech
             feats_len = speech_lengths
-        lfr_factor = max(1, (feats.size()[-1]//80)-1)
+        lfr_factor = max(1, (feats.size()[-1] // 80) - 1)
         batch = {"speech": feats, "speech_lengths": feats_len}
 
         # a. To device
@@ -229,7 +228,8 @@ class Speech2Text:
         enc_len_batch_total = torch.sum(enc_len).item() * self.encoder_downsampling_factor
 
         predictor_outs = self.asr_model.calc_predictor(enc, enc_len)
-        pre_acoustic_embeds, pre_token_length, alphas, pre_peak_index = predictor_outs[0], predictor_outs[1], predictor_outs[2], predictor_outs[3]
+        pre_acoustic_embeds, pre_token_length, alphas, pre_peak_index = predictor_outs[0], predictor_outs[1], \
+                                                                        predictor_outs[2], predictor_outs[3]
         pre_token_length = pre_token_length.round().long()
         if torch.max(pre_token_length) < 1:
             return []
@@ -249,7 +249,7 @@ class Speech2Text:
                 nbest_hyps = self.beam_search(
                     x=x, am_scores=am_scores, maxlenratio=self.maxlenratio, minlenratio=self.minlenratio
                 )
-    
+
                 nbest_hyps = nbest_hyps[: self.nbest]
             else:
                 yseq = am_scores.argmax(dim=-1)
@@ -260,23 +260,23 @@ class Speech2Text:
                     [self.asr_model.sos] + yseq.tolist() + [self.asr_model.eos], device=yseq.device
                 )
                 nbest_hyps = [Hypothesis(yseq=yseq, score=score)]
-                
+
             for hyp in nbest_hyps:
                 assert isinstance(hyp, (Hypothesis)), type(hyp)
-    
+
                 # remove sos/eos and get results
                 last_pos = -1
                 if isinstance(hyp.yseq, list):
                     token_int = hyp.yseq[1:last_pos]
                 else:
                     token_int = hyp.yseq[1:last_pos].tolist()
-    
+
                 # remove blank symbol id, which is assumed to be 0
                 token_int = list(filter(lambda x: x != 0 and x != 2, token_int))
-    
+
                 # Change integer-ids to tokens
                 token = self.converter.ids2tokens(token_int)
-    
+
                 if self.tokenizer is not None:
                     text = self.tokenizer.tokens2text(token)
                 else:
@@ -286,12 +286,14 @@ class Speech2Text:
                     timestamp = time_stamp_lfr6_pl(us_alphas[i], us_cif_peak[i], copy.copy(token), begin_time, end_time)
                     results.append((text, token, token_int, timestamp, enc_len_batch_total, lfr_factor))
                 else:
-                    time_stamp = time_stamp_lfr6(alphas[i:i + 1, ], enc_len[i:i + 1, ], copy.copy(token), begin_time, end_time)
+                    time_stamp = time_stamp_lfr6(alphas[i:i + 1, ], enc_len[i:i + 1, ], copy.copy(token), begin_time,
+                                                 end_time)
                     results.append((text, token, token_int, time_stamp, enc_len_batch_total, lfr_factor))
 
         # assert check_return_type(results)
         return results
 
+
 class Speech2VadSegment:
     """Speech2VadSegment class
 
@@ -333,6 +335,7 @@ class Speech2VadSegment:
         self.device = device
         self.dtype = dtype
         self.frontend = frontend
+        self.batch_size = batch_size
 
     @torch.no_grad()
     def __call__(
@@ -361,56 +364,69 @@ class Speech2VadSegment:
             feats_len = feats_len.int()
         else:
             raise Exception("Need to extract feats first, please configure frontend configuration")
-        batch = {"feats": feats, "feats_lengths": feats_len, "waveform": speech}
-
-        # a. To device
-        batch = to_device(batch, device=self.device)
 
-        # b. Forward Encoder
-        segments = self.vad_model(**batch)
+        # b. Forward Encoder streaming
+        t_offset = 0
+        step = min(feats_len, 6000)
+        segments = [[]] * self.batch_size
+        for t_offset in range(0, feats_len, min(step, feats_len - t_offset)):
+            if t_offset + step >= feats_len - 1:
+                step = feats_len - t_offset
+                is_final_send = True
+            else:
+                is_final_send = False
+            batch = {
+                "feats": feats[:, t_offset:t_offset + step, :],
+                "waveform": speech[:, t_offset * 160:min(speech.shape[-1], (t_offset + step - 1) * 160 + 400)],
+                "is_final_send": is_final_send
+            }
+            # a. To device
+            batch = to_device(batch, device=self.device)
+            segments_part = self.vad_model(**batch)
+            if segments_part:
+                for batch_num in range(0, self.batch_size):
+                    segments[batch_num] += segments_part[batch_num]
 
         return fbanks, segments
 
 
-
 def inference(
-    maxlenratio: float,
-    minlenratio: float,
-    batch_size: int,
-    beam_size: int,
-    ngpu: int,
-    ctc_weight: float,
-    lm_weight: float,
-    penalty: float,
-    log_level: Union[int, str],
-    data_path_and_name_and_type,
-    asr_train_config: Optional[str],
-    asr_model_file: Optional[str],
-    cmvn_file: Optional[str] = None,
-    raw_inputs: Union[np.ndarray, torch.Tensor] = None,
-    lm_train_config: Optional[str] = None,
-    lm_file: Optional[str] = None,
-    token_type: Optional[str] = None,
-    key_file: Optional[str] = None,
-    word_lm_train_config: Optional[str] = None,
-    bpemodel: Optional[str] = None,
-    allow_variable_data_keys: bool = False,
-    streaming: bool = False,
-    output_dir: Optional[str] = None,
-    dtype: str = "float32",
-    seed: int = 0,
-    ngram_weight: float = 0.9,
-    nbest: int = 1,
-    num_workers: int = 1,
-    vad_infer_config: Optional[str] = None,
-    vad_model_file: Optional[str] = None,
-    vad_cmvn_file: Optional[str] = None,
-    time_stamp_writer: bool = False,
-    punc_infer_config: Optional[str] = None,
-    punc_model_file: Optional[str] = None,
-    **kwargs,
+        maxlenratio: float,
+        minlenratio: float,
+        batch_size: int,
+        beam_size: int,
+        ngpu: int,
+        ctc_weight: float,
+        lm_weight: float,
+        penalty: float,
+        log_level: Union[int, str],
+        data_path_and_name_and_type,
+        asr_train_config: Optional[str],
+        asr_model_file: Optional[str],
+        cmvn_file: Optional[str] = None,
+        raw_inputs: Union[np.ndarray, torch.Tensor] = None,
+        lm_train_config: Optional[str] = None,
+        lm_file: Optional[str] = None,
+        token_type: Optional[str] = None,
+        key_file: Optional[str] = None,
+        word_lm_train_config: Optional[str] = None,
+        bpemodel: Optional[str] = None,
+        allow_variable_data_keys: bool = False,
+        streaming: bool = False,
+        output_dir: Optional[str] = None,
+        dtype: str = "float32",
+        seed: int = 0,
+        ngram_weight: float = 0.9,
+        nbest: int = 1,
+        num_workers: int = 1,
+        vad_infer_config: Optional[str] = None,
+        vad_model_file: Optional[str] = None,
+        vad_cmvn_file: Optional[str] = None,
+        time_stamp_writer: bool = False,
+        punc_infer_config: Optional[str] = None,
+        punc_model_file: Optional[str] = None,
+        **kwargs,
 ):
-
     inference_pipeline = inference_modelscope(
         maxlenratio=maxlenratio,
         minlenratio=minlenratio,
@@ -449,63 +465,64 @@ def inference(
     )
     return inference_pipeline(data_path_and_name_and_type, raw_inputs)
 
+
 def inference_modelscope(
-    maxlenratio: float,
-    minlenratio: float,
-    batch_size: int,
-    beam_size: int,
-    ngpu: int,
-    ctc_weight: float,
-    lm_weight: float,
-    penalty: float,
-    log_level: Union[int, str],
-    # data_path_and_name_and_type,
-    asr_train_config: Optional[str],
-    asr_model_file: Optional[str],
-    cmvn_file: Optional[str] = None,
-    lm_train_config: Optional[str] = None,
-    lm_file: Optional[str] = None,
-    token_type: Optional[str] = None,
-    key_file: Optional[str] = None,
-    word_lm_train_config: Optional[str] = None,
-    bpemodel: Optional[str] = None,
-    allow_variable_data_keys: bool = False,
-    output_dir: Optional[str] = None,
-    dtype: str = "float32",
-    seed: int = 0,
-    ngram_weight: float = 0.9,
-    nbest: int = 1,
-    num_workers: int = 1,
-    vad_infer_config: Optional[str] = None,
-    vad_model_file: Optional[str] = None,
-    vad_cmvn_file: Optional[str] = None,
-    time_stamp_writer: bool = True,
-    punc_infer_config: Optional[str] = None,
-    punc_model_file: Optional[str] = None,
-    outputs_dict: Optional[bool] = True,
-    param_dict: dict = None,
-    **kwargs,
+        maxlenratio: float,
+        minlenratio: float,
+        batch_size: int,
+        beam_size: int,
+        ngpu: int,
+        ctc_weight: float,
+        lm_weight: float,
+        penalty: float,
+        log_level: Union[int, str],
+        # data_path_and_name_and_type,
+        asr_train_config: Optional[str],
+        asr_model_file: Optional[str],
+        cmvn_file: Optional[str] = None,
+        lm_train_config: Optional[str] = None,
+        lm_file: Optional[str] = None,
+        token_type: Optional[str] = None,
+        key_file: Optional[str] = None,
+        word_lm_train_config: Optional[str] = None,
+        bpemodel: Optional[str] = None,
+        allow_variable_data_keys: bool = False,
+        output_dir: Optional[str] = None,
+        dtype: str = "float32",
+        seed: int = 0,
+        ngram_weight: float = 0.9,
+        nbest: int = 1,
+        num_workers: int = 1,
+        vad_infer_config: Optional[str] = None,
+        vad_model_file: Optional[str] = None,
+        vad_cmvn_file: Optional[str] = None,
+        time_stamp_writer: bool = True,
+        punc_infer_config: Optional[str] = None,
+        punc_model_file: Optional[str] = None,
+        outputs_dict: Optional[bool] = True,
+        param_dict: dict = None,
+        **kwargs,
 ):
     assert check_argument_types()
-    
+
     if word_lm_train_config is not None:
         raise NotImplementedError("Word LM is not implemented")
     if ngpu > 1:
         raise NotImplementedError("only single GPU decoding is supported")
-    
+
     logging.basicConfig(
         level=log_level,
         format="%(asctime)s (%(module)s:%(lineno)d) %(levelname)s: %(message)s",
     )
-    
+
     if ngpu >= 1 and torch.cuda.is_available():
         device = "cuda"
     else:
         device = "cpu"
-    
+
     # 1. Set random-seed
     set_all_random_seed(seed)
-    
+
     # 2. Build speech2vadsegment
     speech2vadsegment_kwargs = dict(
         vad_infer_config=vad_infer_config,
@@ -516,7 +533,7 @@ def inference_modelscope(
     )
     # logging.info("speech2vadsegment_kwargs: {}".format(speech2vadsegment_kwargs))
     speech2vadsegment = Speech2VadSegment(**speech2vadsegment_kwargs)
-    
+
     # 3. Build speech2text
     speech2text_kwargs = dict(
         asr_train_config=asr_train_config,
@@ -539,14 +556,14 @@ def inference_modelscope(
     )
     speech2text = Speech2Text(**speech2text_kwargs)
     text2punc = None
-    if punc_model_file is not None: 
+    if punc_model_file is not None:
         text2punc = Text2Punc(punc_infer_config, punc_model_file, device=device, dtype=dtype)
 
     if output_dir is not None:
         writer = DatadirWriter(output_dir)
         ibest_writer = writer[f"1best_recog"]
         ibest_writer["token_list"][""] = " ".join(speech2text.asr_train_args.token_list)
-    
+
     def _forward(data_path_and_name_and_type,
                  raw_inputs: Union[np.ndarray, torch.Tensor] = None,
                  output_dir_v2: Optional[str] = None,
@@ -575,7 +592,7 @@ def inference_modelscope(
             use_timestamp = param_dict.get('use_timestamp', True)
         else:
             use_timestamp = True
-    
+
         finish_count = 0
         file_count = 1
         lfr_factor = 6
@@ -586,13 +603,13 @@ def inference_modelscope(
         if output_path is not None:
             writer = DatadirWriter(output_path)
             ibest_writer = writer[f"1best_recog"]
-    
+
         for keys, batch in loader:
             assert isinstance(batch, dict), type(batch)
             assert all(isinstance(s, str) for s in keys), keys
             _bs = len(next(iter(batch.values())))
             assert len(keys) == _bs, f"{len(keys)} != {_bs}"
-    
+
             vad_results = speech2vadsegment(**batch)
             fbanks, vadsegments = vad_results[0], vad_results[1]
             for i, segments in enumerate(vadsegments):
@@ -606,19 +623,20 @@ def inference_modelscope(
                     results = speech2text(**batch)
                     if len(results) < 1:
                         continue
-    
+
                     result_cur = [results[0][:-2]]
                     if j == 0:
                         result_segments = result_cur
                     else:
-                        result_segments = [[result_segments[0][i] + result_cur[0][i] for i in range(len(result_cur[0]))]]
-    
+                        result_segments = [
+                            [result_segments[0][i] + result_cur[0][i] for i in range(len(result_cur[0]))]]
+
                 key = keys[0]
                 result = result_segments[0]
                 text, token, token_int = result[0], result[1], result[2]
                 time_stamp = None if len(result) < 4 else result[3]
-   
-                if use_timestamp and time_stamp is not None: 
+
+                if use_timestamp and time_stamp is not None:
                     postprocessed_result = postprocess_utils.sentence_postprocess(token, time_stamp)
                 else:
                     postprocessed_result = postprocess_utils.sentence_postprocess(token)
@@ -635,13 +653,13 @@ def inference_modelscope(
                 text_postprocessed_punc = text_postprocessed
                 if len(word_lists) > 0 and text2punc is not None:
                     text_postprocessed_punc, punc_id_list = text2punc(word_lists, 20)
-    
+
                 item = {'key': key, 'value': text_postprocessed_punc}
                 if text_postprocessed != "":
                     item['text_postprocessed'] = text_postprocessed
                 if time_stamp_postprocessed != "":
                     item['time_stamp'] = time_stamp_postprocessed
-    
+
                 asr_result_list.append(item)
                 finish_count += 1
                 # asr_utils.print_progress(finish_count / file_count)
@@ -654,11 +672,13 @@ def inference_modelscope(
                     ibest_writer["text_with_punc"][key] = text_postprocessed_punc
                     if time_stamp_postprocessed is not None:
                         ibest_writer["time_stamp"][key] = "{}".format(time_stamp_postprocessed)
-    
+
                 logging.info("decoding, utt: {}, predictions: {}".format(key, text_postprocessed_punc))
         return asr_result_list
+
     return _forward
 
+
 def get_parser():
     parser = config_argparse.ArgumentParser(
         description="ASR Decoding",

funasr/bin/vad_inference.py

@@ -107,10 +107,8 @@ class Speech2VadSegment:
             feats_len = feats_len.int()
         else:
             raise Exception("Need to extract feats first, please configure frontend configuration")
-        # batch = {"feats": feats, "waveform": speech, "is_final_send": True}
-        # segments = self.vad_model(**batch)
 
-        # b. Forward Encoder sreaming
+        # b. Forward Encoder streaming
         t_offset = 0
         step = min(feats_len, 6000)
         segments = [[]] * self.batch_size

vad_inference.py

@@ -1,364 +0,0 @@
-import argparse
-import logging
-import sys
-import json
-from pathlib import Path
-from typing import Any
-from typing import List
-from typing import Optional
-from typing import Sequence
-from typing import Tuple
-from typing import Union
-from typing import Dict
-
-import numpy as np
-import torch
-from typeguard import check_argument_types
-from typeguard import check_return_type
-
-from funasr.fileio.datadir_writer import DatadirWriter
-from funasr.modules.scorers.scorer_interface import BatchScorerInterface
-from funasr.modules.subsampling import TooShortUttError
-from funasr.tasks.vad import VADTask
-from funasr.torch_utils.device_funcs import to_device
-from funasr.torch_utils.set_all_random_seed import set_all_random_seed
-from funasr.utils import config_argparse
-from funasr.utils.cli_utils import get_commandline_args
-from funasr.utils.types import str2bool
-from funasr.utils.types import str2triple_str
-from funasr.utils.types import str_or_none
-from funasr.utils import asr_utils, wav_utils, postprocess_utils
-from funasr.models.frontend.wav_frontend import WavFrontend
-
-header_colors = '\033[95m'
-end_colors = '\033[0m'
-
-global_asr_language: str = 'zh-cn'
-global_sample_rate: Union[int, Dict[Any, int]] = {
-    'audio_fs': 16000,
-    'model_fs': 16000
-}
-
-
-class Speech2VadSegment:
-    """Speech2VadSegment class
-
-    Examples:
-        >>> import soundfile
-        >>> speech2segment = Speech2VadSegment("vad_config.yml", "vad.pt")
-        >>> audio, rate = soundfile.read("speech.wav")
-        >>> speech2segment(audio)
-        [[10, 230], [245, 450], ...]
-
-    """
-
-    def __init__(
-            self,
-            vad_infer_config: Union[Path, str] = None,
-            vad_model_file: Union[Path, str] = None,
-            vad_cmvn_file: Union[Path, str] = None,
-            device: str = "cpu",
-            batch_size: int = 1,
-            dtype: str = "float32",
-            **kwargs,
-    ):
-        assert check_argument_types()
-
-        # 1. Build vad model
-        vad_model, vad_infer_args = VADTask.build_model_from_file(
-            vad_infer_config, vad_model_file, device
-        )
-        frontend = None
-        if vad_infer_args.frontend is not None:
-            frontend = WavFrontend(cmvn_file=vad_cmvn_file, **vad_infer_args.frontend_conf)
-
-        logging.info("vad_model: {}".format(vad_model))
-        logging.info("vad_infer_args: {}".format(vad_infer_args))
-        vad_model.to(dtype=getattr(torch, dtype)).eval()
-
-        self.vad_model = vad_model
-        self.vad_infer_args = vad_infer_args
-        self.device = device
-        self.dtype = dtype
-        self.frontend = frontend
-        self.batch_size = batch_size
-
-    @torch.no_grad()
-    def __call__(
-            self, speech: Union[torch.Tensor, np.ndarray], speech_lengths: Union[torch.Tensor, np.ndarray] = None
-    ) -> List[List[int]]:
-        """Inference
-
-        Args:
-            speech: Input speech data
-        Returns:
-            text, token, token_int, hyp
-
-        """
-        assert check_argument_types()
-
-        # Input as audio signal
-        if isinstance(speech, np.ndarray):
-            speech = torch.tensor(speech)
-
-        if self.frontend is not None:
-            feats, feats_len = self.frontend.forward(speech, speech_lengths)
-            feats = to_device(feats, device=self.device)
-            feats_len = feats_len.int()
-        else:
-            raise Exception("Need to extract feats first, please configure frontend configuration")
-
-        # b. Forward Encoder streaming
-        t_offset = 0
-        step = min(feats_len, 6000)
-        segments = [[]] * self.batch_size
-        for t_offset in range(0, feats_len, min(step, feats_len - t_offset)):
-            if t_offset + step >= feats_len - 1:
-                step = feats_len - t_offset
-                is_final_send = True
-            else:
-                is_final_send = False
-            batch = {
-                "feats": feats[:, t_offset:t_offset + step, :],
-                "waveform": speech[:, t_offset * 160:min(speech.shape[-1], (t_offset + step - 1) * 160 + 400)],
-                "is_final_send": is_final_send
-            }
-            # a. To device
-            batch = to_device(batch, device=self.device)
-            segments_part = self.vad_model(**batch)
-            if segments_part:
-                for batch_num in range(0, self.batch_size):
-                    segments[batch_num] += segments_part[batch_num]
-        return segments
-
-
-def inference(
-        batch_size: int,
-        ngpu: int,
-        log_level: Union[int, str],
-        data_path_and_name_and_type,
-        vad_infer_config: Optional[str],
-        vad_model_file: Optional[str],
-        vad_cmvn_file: Optional[str] = None,
-        raw_inputs: Union[np.ndarray, torch.Tensor] = None,
-        key_file: Optional[str] = None,
-        allow_variable_data_keys: bool = False,
-        output_dir: Optional[str] = None,
-        dtype: str = "float32",
-        seed: int = 0,
-        num_workers: int = 1,
-        **kwargs,
-):
-    inference_pipeline = inference_modelscope(
-        batch_size=batch_size,
-        ngpu=ngpu,
-        log_level=log_level,
-        vad_infer_config=vad_infer_config,
-        vad_model_file=vad_model_file,
-        vad_cmvn_file=vad_cmvn_file,
-        key_file=key_file,
-        allow_variable_data_keys=allow_variable_data_keys,
-        output_dir=output_dir,
-        dtype=dtype,
-        seed=seed,
-        num_workers=num_workers,
-        **kwargs,
-    )
-    return inference_pipeline(data_path_and_name_and_type, raw_inputs)
-
-
-def inference_modelscope(
-        batch_size: int,
-        ngpu: int,
-        log_level: Union[int, str],
-        # data_path_and_name_and_type,
-        vad_infer_config: Optional[str],
-        vad_model_file: Optional[str],
-        vad_cmvn_file: Optional[str] = None,
-        # raw_inputs: Union[np.ndarray, torch.Tensor] = None,
-        key_file: Optional[str] = None,
-        allow_variable_data_keys: bool = False,
-        output_dir: Optional[str] = None,
-        dtype: str = "float32",
-        seed: int = 0,
-        num_workers: int = 1,
-        **kwargs,
-):
-    assert check_argument_types()
-    if batch_size > 1:
-        raise NotImplementedError("batch decoding is not implemented")
-    if ngpu > 1:
-        raise NotImplementedError("only single GPU decoding is supported")
-
-    logging.basicConfig(
-        level=log_level,
-        format="%(asctime)s (%(module)s:%(lineno)d) %(levelname)s: %(message)s",
-    )
-
-    if ngpu >= 1 and torch.cuda.is_available():
-        device = "cuda"
-    else:
-        device = "cpu"
-
-    # 1. Set random-seed
-    set_all_random_seed(seed)
-
-    # 2. Build speech2vadsegment
-    speech2vadsegment_kwargs = dict(
-        vad_infer_config=vad_infer_config,
-        vad_model_file=vad_model_file,
-        vad_cmvn_file=vad_cmvn_file,
-        device=device,
-        dtype=dtype,
-    )
-    logging.info("speech2vadsegment_kwargs: {}".format(speech2vadsegment_kwargs))
-    speech2vadsegment = Speech2VadSegment(**speech2vadsegment_kwargs)
-
-    def _forward(
-            data_path_and_name_and_type,
-            raw_inputs: Union[np.ndarray, torch.Tensor] = None,
-            output_dir_v2: Optional[str] = None,
-            fs: dict = None,
-            param_dict: dict = None,
-    ):
-        # 3. Build data-iterator
-        loader = VADTask.build_streaming_iterator(
-            data_path_and_name_and_type,
-            dtype=dtype,
-            batch_size=batch_size,
-            key_file=key_file,
-            num_workers=num_workers,
-            preprocess_fn=VADTask.build_preprocess_fn(speech2vadsegment.vad_infer_args, False),
-            collate_fn=VADTask.build_collate_fn(speech2vadsegment.vad_infer_args, False),
-            allow_variable_data_keys=allow_variable_data_keys,
-            inference=True,
-        )
-
-        finish_count = 0
-        file_count = 1
-        # 7 .Start for-loop
-        # FIXME(kamo): The output format should be discussed about
-        output_path = output_dir_v2 if output_dir_v2 is not None else output_dir
-        if output_path is not None:
-            writer = DatadirWriter(output_path)
-            ibest_writer = writer[f"1best_recog"]
-        else:
-            writer = None
-            ibest_writer = None
-
-        vad_results = []
-        for keys, batch in loader:
-            assert isinstance(batch, dict), type(batch)
-            assert all(isinstance(s, str) for s in keys), keys
-            _bs = len(next(iter(batch.values())))
-            assert len(keys) == _bs, f"{len(keys)} != {_bs}"
-
-            # do vad segment
-            results = speech2vadsegment(**batch)
-            for i, _ in enumerate(keys):
-                results[i] = json.dumps(results[i])
-                item = {'key': keys[i], 'value': results[i]}
-                vad_results.append(item)
-                if writer is not None:
-                    results[i] = json.loads(results[i])
-                    ibest_writer["text"][keys[i]] = "{}".format(results[i])
-
-        return vad_results
-
-    return _forward
-
-
-def get_parser():
-    parser = config_argparse.ArgumentParser(
-        description="VAD Decoding",
-        formatter_class=argparse.ArgumentDefaultsHelpFormatter,
-    )
-
-    # Note(kamo): Use '_' instead of '-' as separator.
-    # '-' is confusing if written in yaml.
-    parser.add_argument(
-        "--log_level",
-        type=lambda x: x.upper(),
-        default="INFO",
-        choices=("CRITICAL", "ERROR", "WARNING", "INFO", "DEBUG", "NOTSET"),
-        help="The verbose level of logging",
-    )
-
-    parser.add_argument("--output_dir", type=str, required=False)
-    parser.add_argument(
-        "--ngpu",
-        type=int,
-        default=0,
-        help="The number of gpus. 0 indicates CPU mode",
-    )
-    parser.add_argument(
-        "--gpuid_list",
-        type=str,
-        default="",
-        help="The visible gpus",
-    )
-    parser.add_argument("--seed", type=int, default=0, help="Random seed")
-    parser.add_argument(
-        "--dtype",
-        default="float32",
-        choices=["float16", "float32", "float64"],
-        help="Data type",
-    )
-    parser.add_argument(
-        "--num_workers",
-        type=int,
-        default=1,
-        help="The number of workers used for DataLoader",
-    )
-
-    group = parser.add_argument_group("Input data related")
-    group.add_argument(
-        "--data_path_and_name_and_type",
-        type=str2triple_str,
-        required=False,
-        action="append",
-    )
-    group.add_argument("--raw_inputs", type=list, default=None)
-    # example=[{'key':'EdevDEWdIYQ_0021','file':'/mnt/data/jiangyu.xzy/test_data/speech_io/SPEECHIO_ASR_ZH00007_zhibodaihuo/wav/EdevDEWdIYQ_0021.wav'}])
-    group.add_argument("--key_file", type=str_or_none)
-    group.add_argument("--allow_variable_data_keys", type=str2bool, default=False)
-
-    group = parser.add_argument_group("The model configuration related")
-    group.add_argument(
-        "--vad_infer_config",
-        type=str,
-        help="VAD infer configuration",
-    )
-    group.add_argument(
-        "--vad_model_file",
-        type=str,
-        help="VAD model parameter file",
-    )
-    group.add_argument(
-        "--vad_cmvn_file",
-        type=str,
-        help="Global cmvn file",
-    )
-
-    group = parser.add_argument_group("infer related")
-    group.add_argument(
-        "--batch_size",
-        type=int,
-        default=1,
-        help="The batch size for inference",
-    )
-
-    return parser
-
-
-def main(cmd=None):
-    print(get_commandline_args(), file=sys.stderr)
-    parser = get_parser()
-    args = parser.parse_args(cmd)
-    kwargs = vars(args)
-    kwargs.pop("config", None)
-    inference(**kwargs)
-
-
-if __name__ == "__main__":
-    main()