support asr_inference_paraformer_vad_punc

凌匀

2023-02-16 8689fb676d6cd28894f55c8cac43409e7bf4cd38

 funasr/bin/asr_inference_paraformer_vad_punc.py

@@ -14,6 +14,7 @@
from typing import Any
from typing import List
import math
import copy
import numpy as np
import torch
from typeguard import check_argument_types
@@ -38,20 +39,13 @@
from funasr.utils import asr_utils, wav_utils, postprocess_utils
from funasr.models.frontend.wav_frontend import WavFrontend
from funasr.tasks.vad import VADTask
from funasr.utils.timestamp_tools import time_stamp_lfr6
from funasr.utils.timestamp_tools import time_stamp_lfr6, time_stamp_lfr6_pl
from funasr.bin.punctuation_infer import Text2Punc
from funasr.torch_utils.forward_adaptor import ForwardAdaptor
from funasr.datasets.preprocessor import CommonPreprocessor
from funasr.punctuation.text_preprocessor import split_to_mini_sentence
from funasr.models.e2e_asr_paraformer import BiCifParaformer

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 Speech2Text:
    """Speech2Text class
@@ -150,7 +144,7 @@
        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
@@ -190,12 +184,11 @@
        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

@@ -221,7 +214,7 @@
        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
@@ -235,12 +228,17 @@
        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 []
        decoder_outs = self.asr_model.cal_decoder_with_predictor(enc, enc_len, pre_acoustic_embeds, pre_token_length)
        decoder_out, ys_pad_lens = decoder_outs[0], decoder_outs[1]

        if isinstance(self.asr_model, BiCifParaformer):
            _, _, us_alphas, us_cif_peak = self.asr_model.calc_predictor_timestamp(enc, enc_len,
                                                                                   pre_token_length)  # test no bias cif2

        results = []
        b, n, d = decoder_out.size()
@@ -251,7 +249,7 @@
                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)
@@ -262,34 +260,39 @@
                    [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:
                    text = None

                time_stamp = time_stamp_lfr6(alphas[i:i+1,], enc_len[i:i+1,], token, begin_time, end_time)
    
                results.append((text, token, token_int, time_stamp, enc_len_batch_total, lfr_factor))
                if isinstance(self.asr_model, BiCifParaformer):
                    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)
                    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
@@ -332,6 +335,7 @@
        self.device = device
        self.dtype = dtype
        self.frontend = frontend
        self.batch_size = batch_size

    @torch.no_grad()
    def __call__(
@@ -360,56 +364,69 @@
            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,
@@ -448,63 +465,64 @@
    )
    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,
@@ -515,7 +533,7 @@
    )
    # logging.info("speech2vadsegment_kwargs: {}".format(speech2vadsegment_kwargs))
    speech2vadsegment = Speech2VadSegment(**speech2vadsegment_kwargs)
    

    # 3. Build speech2text
    speech2text_kwargs = dict(
        asr_train_config=asr_train_config,
@@ -537,17 +555,19 @@
        nbest=nbest,
    )
    speech2text = Speech2Text(**speech2text_kwargs)
    
    text2punc = Text2Punc(punc_infer_config, punc_model_file, device=device, dtype=dtype)
    text2punc = 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,
                 fs: dict = None,
                 param_dict: dict = None,
                 ):
        # 3. Build data-iterator
@@ -558,6 +578,7 @@
        loader = ASRTask.build_streaming_iterator(
            data_path_and_name_and_type,
            dtype=dtype,
            fs=fs,
            batch_size=1,
            key_file=key_file,
            num_workers=num_workers,
@@ -566,38 +587,33 @@
            allow_variable_data_keys=allow_variable_data_keys,
            inference=True,
        )
        
        forward_time_total = 0.0
        length_total = 0.0

        if param_dict is not None:
            use_timestamp = param_dict.get('use_timestamp', True)
        else:
            use_timestamp = True

        finish_count = 0
        file_count = 1
        lfr_factor = 6
        # 7 .Start for-loop
        asr_result_list = []
        output_path = output_dir_v2 if output_dir_v2 is not None else output_dir
        writer = None
        if output_path is not None:
            writer = DatadirWriter(output_path)
            ibest_writer = writer[f"1best_recog"]
            # ibest_writer["punc_dict"][""] = " ".join(punc_infer_config.punc_list)
            # ibest_writer["token_list"][""] = " ".join(asr_train_config.token_list)
        else:
            writer = None
        

        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}"
            # batch = {k: v for k, v in batch.items() if not k.endswith("_lengths")}
            
            logging.info("decoding, utt_id: {}".format(keys))
            # N-best list of (text, token, token_int, hyp_object)
            time_beg = time.time()

            vad_results = speech2vadsegment(**batch)
            time_end = time.time()
            fbanks, vadsegments = vad_results[0], vad_results[1]
            for i, segments in enumerate(vadsegments):
                result_segments = [["", [], [], ]]
                result_segments = [["", [], [], []]]
                for j, segment_idx in enumerate(segments):
                    bed_idx, end_idx = int(segment_idx[0] / 10), int(segment_idx[1] / 10)
                    segment = fbanks[:, bed_idx:end_idx, :].to(device)
@@ -606,79 +622,63 @@
                             "end_time": vadsegments[i][j][1]}
                    results = speech2text(**batch)
                    if len(results) < 1:
                        hyp = Hypothesis(score=0.0, scores={}, states={}, yseq=[])
                        results = [[" ", ["sil"], [2], 0, 1, 6]] * nbest
                    time_end = time.time()
                    forward_time = time_end - time_beg
                    lfr_factor = results[0][-1]
                    length = results[0][-2]
                    forward_time_total += forward_time
                    length_total += length
                    logging.info(
                        "decoding, feature length: {}, forward_time: {:.4f}, rtf: {:.4f}".
                        format(length, forward_time, 100 * forward_time / (length * lfr_factor)))
                        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]
                
                # Create a directory: outdir/{n}best_recog

                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)
                text_postprocessed = ""
                time_stamp_postprocessed = ""
                text_postprocessed_punc = postprocessed_result
                if len(postprocessed_result) == 3:
                    text_postprocessed, time_stamp_postprocessed, word_lists = postprocessed_result[0], \
                                                                               postprocessed_result[1], \
                                                                               postprocessed_result[2]
                else:
                    text_postprocessed, word_lists = postprocessed_result[0], postprocessed_result[1]

                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)
                if writer is not None:
                    # Write the result to each file
                    ibest_writer["token"][key] = " ".join(token)
                    ibest_writer["token_int"][key] = " ".join(map(str, token_int))
                    ibest_writer["vad"][key] = "{}".format(vadsegments)
                
                if text is not None:
                    postprocessed_result = postprocess_utils.sentence_postprocess(token, time_stamp)
                    if len(postprocessed_result) == 3:
                        text_postprocessed, time_stamp_postprocessed, word_lists = postprocessed_result[0], \
                                                                                   postprocessed_result[1], \
                                                                                   postprocessed_result[2]
                        if len(word_lists) > 0: 
                            text_postprocessed_punc, punc_id_list = text2punc(word_lists, 20)
                            text_postprocessed_punc_time_stamp = json.dumps({"predictions": text_postprocessed_punc,
                                                                             "time_stamp": time_stamp_postprocessed},
                                                                            ensure_ascii=False)
                        else:
                            text_postprocessed_punc = ""
                            punc_id_list = []
                            text_postprocessed_punc_time_stamp = ""
                            
                    else:
                        text_postprocessed = ""
                        time_stamp_postprocessed = ""
                        word_lists = ""
                        text_postprocessed_punc_time_stamp = ""
                        punc_id_list = ""
                        text_postprocessed_punc = ""
                    ibest_writer["text"][key] = text_postprocessed
                    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)

                    item = {'key': key, 'value': text_postprocessed_punc, 'text_postprocessed': text_postprocessed,
                            'time_stamp': time_stamp_postprocessed, 'token': token}
                    asr_result_list.append(item)
                    finish_count += 1
                    # asr_utils.print_progress(finish_count / file_count)
                    if writer is not None:
                        ibest_writer["text"][key] = text_postprocessed
                        ibest_writer["punc_id"][key] = "{}".format(punc_id_list)
                        ibest_writer["text_with_punc"][key] = text_postprocessed_punc_time_stamp
                        if time_stamp_postprocessed is not None:
                            ibest_writer["time_stamp"][key] = "{}".format(time_stamp_postprocessed)
                
                logging.info("decoding, utt: {}, predictions: {}, time_stamp: {}".format(key, text_postprocessed_punc,
                                                                                         time_stamp_postprocessed))
        
        logging.info("decoding, feature length total: {}, forward_time total: {:.4f}, rtf avg: {:.4f}".
                     format(length_total, forward_time_total, 100 * forward_time_total / (length_total * lfr_factor+1e-6)))
                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",