AttributeError: 'SoundFile' object has no attribute 'frames'


[NeMo I 2023-10-17 17:09:36 speaker_utils:93] Number of files to diarize: 1
[NeMo I 2023-10-17 17:09:36 clustering_diarizer:309] Split long audio file to avoid CUDA memory issue
splitting manifest: 100%|██████████████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:02<00:00,  2.26s/it]
[NeMo I 2023-10-17 17:09:39 vad_utils:107] The prepared manifest file exists. Overwriting!
[NeMo I 2023-10-17 17:09:39 classification_models:272] Perform streaming frame-level VAD
[NeMo I 2023-10-17 17:09:39 collections:301] Filtered duration for loading collection is  0.00 hours.
[NeMo I 2023-10-17 17:09:39 collections:302] Dataset loaded with 71 items, total duration of  0.98 hours.
[NeMo I 2023-10-17 17:09:39 collections:304] # 71 files loaded accounting to # 1 labels
vad: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 71/71 [00:10<00:00,  6.99it/s]
[NeMo I 2023-10-17 17:09:49 clustering_diarizer:250] Generating predictions with overlapping input segments
[NeMo I 2023-10-17 17:10:12 clustering_diarizer:262] Converting frame level prediction to speech/no-speech segment in start and end times format.            
creating speech segments: 100%|████████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:02<00:00,  2.67s/it]
╭─────────────────────────────── Traceback (most recent call last) ────────────────────────────────╮
│ /mnt/bigdisk/whisper-diarization/nemo_process.py:29 in <module>                                  │
│                                                                                                  │
│   26                                                                                             │
│   27 # Initialize NeMo MSDD diarization model                                                    │
│   28 msdd_model = NeuralDiarizer(cfg=create_config(temp_path)).to(args.device)                   │
│ ❱ 29 msdd_model.diarize()                                                                        │
│   30                                                                                             │
│                                                                                                  │
│ /mnt/bigdisk/miniconda3/lib/python3.10/site-packages/torch/utils/_contextlib.py:115 in           │
│ decorate_context                                                                                 │
│                                                                                                  │
│   112 │   @functools.wraps(func)                                                                 │
│   113 │   def decorate_context(*args, **kwargs):                                                 │
│   114 │   │   with ctx_factory():                                                                │
│ ❱ 115 │   │   │   return func(*args, **kwargs)                                                   │
│   116 │                                                                                          │
│   117 │   return decorate_context                                                                │
│   118                                                                                            │
│                                                                                                  │
│ /mnt/bigdisk/miniconda3/lib/python3.10/site-packages/nemo/collections/asr/models/msdd_models.py: │
│ 1180 in diarize                                                                                  │
│                                                                                                  │
│   1177 │   │   Note that the result of MSDD can include multiple speakers at the same time. The  │
│   1178 │   │   function that can generate overlapping timestamps. `self.run_overlap_aware_eval(  │
│   1179 │   │   """                                                                               │
│ ❱ 1180 │   │   self.clustering_embedding.prepare_cluster_embs_infer()                            │
│   1181 │   │   self.msdd_model.pairwise_infer = True                                             │
│   1182 │   │   self.get_emb_clus_infer(self.clustering_embedding)                                │
│   1183 │   │   preds_list, targets_list, signal_lengths_list = self.run_pairwise_diarization()   │
│                                                                                                  │
│ /mnt/bigdisk/miniconda3/lib/python3.10/site-packages/nemo/collections/asr/models/msdd_models.py: │
│ 699 in prepare_cluster_embs_infer                                                                │
│                                                                                                  │
│    696 │   │   Launch clustering diarizer to prepare embedding vectors and clustering results.   │
│    697 │   │   """                                                                               │
│    698 │   │   self.max_num_speakers = self.cfg_diar_infer.diarizer.clustering.parameters.max_n  │
│ ❱  699 │   │   self.emb_sess_test_dict, self.emb_seq_test, self.clus_test_label_dict, _ = self.  │
│    700 │   │   │   self._cfg_msdd.test_ds.manifest_filepath, self._cfg_msdd.test_ds.emb_dir      │
│    701 │   │   )                                                                                 │
│    702                                                                                           │
│                                                                                                  │
│ /mnt/bigdisk/miniconda3/lib/python3.10/site-packages/nemo/collections/asr/models/msdd_models.py: │
│ 866 in run_clustering_diarizer                                                                   │
│                                                                                                  │
│    863 │   │                                                                                     │
│    864 │   │   logging.info(f"Multiscale Weights: {self.clus_diar_model.multiscale_args_dict['m  │
│    865 │   │   logging.info(f"Clustering Parameters: {clustering_params_str}")                   │
│ ❱  866 │   │   scores = self.clus_diar_model.diarize(batch_size=self.cfg_diar_infer.batch_size)  │
│    867 │   │                                                                                     │
│    868 │   │   # If RTTM (ground-truth diarization annotation) files do not exist, scores is No  │
│    869 │   │   if scores is not None:                                                            │
│                                                                                                  │
│ /mnt/bigdisk/miniconda3/lib/python3.10/site-packages/nemo/collections/asr/models/clustering_diar │
│ izer.py:437 in diarize                                                                           │
│                                                                                                  │
│   434 │   │   os.makedirs(out_rttm_dir, exist_ok=True)                                           │
│   435 │   │                                                                                      │
│   436 │   │   # Speech Activity Detection                                                        │
│ ❱ 437 │   │   self._perform_speech_activity_detection()                                          │
│   438 │   │                                                                                      │
│   439 │   │   # Segmentation                                                                     │
│   440 │   │   scales = self.multiscale_args_dict['scale_dict'].items()                           │
│                                                                                                  │
│ /mnt/bigdisk/miniconda3/lib/python3.10/site-packages/nemo/collections/asr/models/clustering_diar │
│ izer.py:325 in _perform_speech_activity_detection                                                │
│                                                                                                  │
│   322 │   │   │   │   )                                                                          │
│   323 │   │   │                                                                                  │
│   324 │   │   │   self._setup_vad_test_data(manifest_vad_input)                                  │
│ ❱ 325 │   │   │   self._run_vad(manifest_vad_input)                                              │
│   326 │   │                                                                                      │
│   327 │   │   elif self._diarizer_params.vad.external_vad_manifest is not None:                  │
│   328 │   │   │   self._speaker_manifest_path = self._diarizer_params.vad.external_vad_manifes   │
│                                                                                                  │
│ /mnt/bigdisk/miniconda3/lib/python3.10/site-packages/nemo/collections/asr/models/clustering_diar │
│ izer.py:281 in _run_vad                                                                          │
│                                                                                                  │
│   278 │   │   │   else:                                                                          │
│   279 │   │   │   │   logging.warning(f"no vad file found for {key} due to zero or negative du   │
│   280 │   │                                                                                      │
│ ❱ 281 │   │   write_rttm2manifest(AUDIO_VAD_RTTM_MAP, self._vad_out_file)                        │
│   282 │   │   self._speaker_manifest_path = self._vad_out_file                                   │
│   283 │                                                                                          │
│   284 │   def _run_segmentation(self, window: float, shift: float, scale_tag: str = ''):         │
│                                                                                                  │
│ /mnt/bigdisk/miniconda3/lib/python3.10/site-packages/nemo/collections/asr/parts/utils/speaker_ut │
│ ils.py:858 in write_rttm2manifest                                                                │
│                                                                                                  │
│    855 │   │   for uniq_id in AUDIO_RTTM_MAP:                                                    │
│    856 │   │   │   rttm_file_path = AUDIO_RTTM_MAP[uniq_id]['rttm_filepath']                     │
│    857 │   │   │   rttm_lines = read_rttm_lines(rttm_file_path)                                  │
│ ❱  858 │   │   │   offset, duration = get_offset_and_duration(AUDIO_RTTM_MAP, uniq_id, decimals  │
│    859 │   │   │   vad_start_end_list_raw = []                                                   │
│    860 │   │   │   for line in rttm_lines:                                                       │
│    861 │   │   │   │   start, dur = get_vad_out_from_rttm_line(line)                             │
│                                                                                                  │
│ /mnt/bigdisk/miniconda3/lib/python3.10/site-packages/nemo/collections/asr/parts/utils/speaker_ut │
│ ils.py:565 in get_offset_and_duration                                                            │
│                                                                                                  │
│    562 │   │   offset = round(AUDIO_RTTM_MAP[uniq_id]['offset'], decimals)                       │
│    563 │   else:                                                                                 │
│    564 │   │   sound = sf.SoundFile(audio_path)                                                  │
│ ❱  565 │   │   duration = sound.frames / sound.samplerate                                        │
│    566 │   │   offset = 0.0                                                                      │
│    567 │   return offset, duration                                                               │
│    568                                                                                           │
│                                                                                                  │
│ /mnt/bigdisk/miniconda3/lib/python3.10/site-packages/soundfile.py:822 in __getattr__             │
│                                                                                                  │
│    819 │   │   │   data = _snd.sf_get_string(self._file, _str_types[name])                       │
│    820 │   │   │   return _ffi.string(data).decode('utf-8', 'replace') if data else ""           │
│    821 │   │   else:                                                                             │
│ ❱  822 │   │   │   raise AttributeError(                                                         │
│    823 │   │   │   │   "'SoundFile' object has no attribute {0!r}".format(name))                 │
│    824 │                                                                                         │
│    825 │   def __len__(self):                                                                    │
╰──────────────────────────────────────────────────────────────────────────────────────────────────╯
AttributeError: 'SoundFile' object has no attribute 'frames'```
MahmoudAshraf97 / whisper-diarization

AttributeError: 'SoundFile' object has no attribute 'frames' #108
