Merge pull request #177 from alibaba-damo-academy/dev_timestamp

zhifu gao

2023-03-02 94cb66dbb9ae12e044a41fb8a3d84e1835ee7e7b

 funasr/bin/vad_inference.py

@@ -1,6 +1,7 @@
import argparse
import logging
import sys
import json
from pathlib import Path
from typing import Any
from typing import List
@@ -10,6 +11,7 @@
from typing import Union
from typing import Dict

import math
import numpy as np
import torch
from typeguard import check_argument_types
@@ -80,11 +82,13 @@
        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]]:
            self, speech: Union[torch.Tensor, np.ndarray], speech_lengths: Union[torch.Tensor, np.ndarray] = None,
            in_cache: Dict[str, torch.Tensor] = dict()
    ) -> Tuple[List[List[int]], Dict[str, torch.Tensor]]:
        """Inference

        Args:
@@ -100,22 +104,38 @@
            speech = torch.tensor(speech)

        if self.frontend is not None:
            feats, feats_len = self.frontend.forward(speech, speech_lengths)
            self.frontend.filter_length_max = math.inf
            fbanks, fbanks_len = self.frontend.forward_fbank(speech, speech_lengths)
            feats, feats_len = self.frontend.forward_lfr_cmvn(fbanks, fbanks_len)
            fbanks = to_device(fbanks, device=self.device)
            feats = to_device(feats, device=self.device)
            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)

        return segments


        # b. Forward Encoder streaming
        t_offset = 0
        step = min(feats_len.max(), 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 = True
            else:
                is_final = 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": is_final,
                "in_cache": in_cache
            }
            # a. To device
            batch = to_device(batch, device=self.device)
            segments_part, in_cache = 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(
@@ -152,11 +172,12 @@
    )
    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,
        # data_path_and_name_and_type,
        vad_infer_config: Optional[str],
        vad_model_file: Optional[str],
        vad_cmvn_file: Optional[str] = None,
@@ -167,7 +188,6 @@
        dtype: str = "float32",
        seed: int = 0,
        num_workers: int = 1,
        param_dict: dict = None,
        **kwargs,
):
    assert check_argument_types()
@@ -201,13 +221,17 @@
    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,
            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
        if data_path_and_name_and_type is None and raw_inputs is not None:
            if isinstance(raw_inputs, torch.Tensor):
                raw_inputs = raw_inputs.numpy()
            data_path_and_name_and_type = [raw_inputs, "speech", "waveform"]
        loader = VADTask.build_streaming_iterator(
            data_path_and_name_and_type,
            dtype=dtype,
@@ -238,14 +262,15 @@
            assert all(isinstance(s, str) for s in keys), keys
            _bs = len(next(iter(batch.values())))
            assert len(keys) == _bs, f"{len(keys)} != {_bs}"
            # batch = {k: v[0] for k, v in batch.items() if not k.endswith("_lengths")}

            # do vad segment
            results = speech2vadsegment(**batch)
            _, 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