| | |
|---|
| | | fusion_encoder_class = tables.encoder_classes.get(fusion_encoder) |
|---|
| | | fusion_encoder = fusion_encoder_class(**fusion_encoder_conf) |
|---|
| | | bias_predictor_class = tables.encoder_classes.get(bias_predictor) |
|---|
| | | bias_predictor = bias_predictor_class(bias_predictor_conf) |
|---|
| | | bias_predictor = bias_predictor_class(**bias_predictor_conf) |
|---|
| | | |
|---|
| | | if decoder is not None: |
|---|
| | | decoder_class = tables.decoder_classes.get(decoder) |
|---|
| | |
|---|
| | | logging.info("enable beam_search") |
|---|
| | | self.init_beam_search(**kwargs) |
|---|
| | | self.nbest = kwargs.get("nbest", 1) |
|---|
| | | pdb.set_trace() |
|---|
| | | |
|---|
| | | meta_data = {} |
|---|
| | | if isinstance(data_in, torch.Tensor) and kwargs.get("data_type", "sound") == "fbank": # fbank |
|---|
| | |
|---|
| | | else: |
|---|
| | | # extract fbank feats |
|---|
| | | time1 = time.perf_counter() |
|---|
| | | pdb.set_trace() |
|---|
| | | audio_sample_list = load_audio_text_image_video(data_in, fs=frontend.fs, audio_fs=kwargs.get("fs", 16000), |
|---|
| | | data_type=kwargs.get("data_type", "sound"), |
|---|
| | | tokenizer=tokenizer) |
|---|
| | | time2 = time.perf_counter() |
|---|
| | | meta_data["load_data"] = f"{time2 - time1:0.3f}" |
|---|
| | | pdb.set_trace() |
|---|
| | | speech, speech_lengths = extract_fbank(audio_sample_list, data_type=kwargs.get("data_type", "sound"), |
|---|
| | | frontend=frontend) |
|---|
| | | time3 = time.perf_counter() |
|---|
