update m2met2 docs

yhliang

2023-04-27 32d2b3ec153e53176da710ebcc0aba5669effd8a

 funasr/runtime/python/onnxruntime/funasr_onnx/utils/e2e_vad.py

@@ -1,3 +1,7 @@
# -*- encoding: utf-8 -*-
# Copyright FunASR (https://github.com/alibaba-damo-academy/FunASR). All Rights Reserved.
#  MIT License  (https://opensource.org/licenses/MIT)

from enum import Enum
from typing import List, Tuple, Dict, Any

@@ -189,6 +193,11 @@


class E2EVadModel():
    """
    Author: Speech Lab of DAMO Academy, Alibaba Group
    Deep-FSMN for Large Vocabulary Continuous Speech Recognition
    https://arxiv.org/abs/1803.05030
    """
    def __init__(self, vad_post_args: Dict[str, Any]):
        super(E2EVadModel, self).__init__()
        self.vad_opts = VADXOptions(**vad_post_args)
@@ -439,10 +448,9 @@
                        - 1)) / self.vad_opts.noise_frame_num_used_for_snr

        return frame_state
     

    def __call__(self, score: np.ndarray, waveform: np.ndarray,
                is_final: bool = False, max_end_sil: int = 800
                is_final: bool = False, max_end_sil: int = 800, online: bool = False
                ):
        self.max_end_sil_frame_cnt_thresh = max_end_sil - self.vad_opts.speech_to_sil_time_thres
        self.waveform = waveform  # compute decibel for each frame
@@ -457,20 +465,29 @@
            segment_batch = []
            if len(self.output_data_buf) > 0:
                for i in range(self.output_data_buf_offset, len(self.output_data_buf)):
                    if not self.output_data_buf[i].contain_seg_start_point:
                        continue
                    if not self.next_seg and not self.output_data_buf[i].contain_seg_end_point:
                        continue
                    start_ms = self.output_data_buf[i].start_ms if self.next_seg else -1
                    if self.output_data_buf[i].contain_seg_end_point:
                        end_ms = self.output_data_buf[i].end_ms
                        self.next_seg = True
                        self.output_data_buf_offset += 1
                    if online:
                        if not self.output_data_buf[i].contain_seg_start_point:
                            continue
                        if not self.next_seg and not self.output_data_buf[i].contain_seg_end_point:
                            continue
                        start_ms = self.output_data_buf[i].start_ms if self.next_seg else -1
                        if self.output_data_buf[i].contain_seg_end_point:
                            end_ms = self.output_data_buf[i].end_ms
                            self.next_seg = True
                            self.output_data_buf_offset += 1
                        else:
                            end_ms = -1
                            self.next_seg = False
                    else:
                        end_ms = -1
                        self.next_seg = False
                        if not is_final and (not self.output_data_buf[i].contain_seg_start_point or not self.output_data_buf[
                            i].contain_seg_end_point):
                            continue
                        start_ms = self.output_data_buf[i].start_ms
                        end_ms = self.output_data_buf[i].end_ms
                        self.output_data_buf_offset += 1
                    segment = [start_ms, end_ms]
                    segment_batch.append(segment)

            if segment_batch:
                segments.append(segment_batch)
        if is_final:
@@ -605,3 +622,4 @@
        if self.vad_state_machine == VadStateMachine.kVadInStateEndPointDetected and \
                self.vad_opts.detect_mode == VadDetectMode.kVadMutipleUtteranceDetectMode.value:
            self.ResetDetection()