Resolve merge conflict

游雁

2024-01-14 8912e0696af069de47646fdb8a9d9c4e086e88b3

 funasr/models/bicif_paraformer/model.py

@@ -23,7 +23,7 @@

from funasr.models.paraformer.search import Hypothesis

from funasr.utils.load_utils import load_audio_and_text_image_video, extract_fbank
from funasr.utils.load_utils import load_audio_text_image_video, extract_fbank
from funasr.utils import postprocess_utils
from funasr.utils.datadir_writer import DatadirWriter
from funasr.utils.timestamp_tools import ts_prediction_lfr6_standard
@@ -243,7 +243,7 @@
        else:
            # extract fbank feats
            time1 = time.perf_counter()
            audio_sample_list = load_audio_and_text_image_video(data_in, fs=frontend.fs, audio_fs=kwargs.get("fs", 16000))
            audio_sample_list = load_audio_text_image_video(data_in, fs=frontend.fs, audio_fs=kwargs.get("fs", 16000))
            time2 = time.perf_counter()
            meta_data["load_data"] = f"{time2 - time1:0.3f}"
            speech, speech_lengths = extract_fbank(audio_sample_list, data_type=kwargs.get("data_type", "sound"),
@@ -252,7 +252,8 @@
            meta_data["extract_feat"] = f"{time3 - time2:0.3f}"
            meta_data["batch_data_time"] = speech_lengths.sum().item() * frontend.frame_shift * frontend.lfr_n / 1000
        
        speech.to(device=kwargs["device"]), speech_lengths.to(device=kwargs["device"])
        speech = speech.to(device=kwargs["device"])
        speech_lengths = speech_lengths.to(device=kwargs["device"])
        
        # Encoder
        encoder_out, encoder_out_lens = self.encode(speech, speech_lengths)