change the vector csv.spk_id to csv.label, test=doc

examples/voxceleb/sv0/local/make_rirs_noise_csv_dataset_from_json.py

@@ -25,7 +25,7 @@ from yacs.config import CfgNode
 
 from paddleaudio import load as load_audio
 from paddlespeech.s2t.utils.log import Log
-from paddlespeech.vector.utils.utils import get_chunks
+from paddlespeech.vector.utils.vector_utils import get_chunks
 
 logger = Log(__name__).getlog()
 
@@ -57,7 +57,9 @@ def get_chunks_list(wav_file: str,
             end_sample = int(float(e) * sr)
 
             # currently, all vector csv data format use one representation
-            # id, duration, wav, start, stop, spk_id
+            # id, duration, wav, start, stop, label
+            # in rirs noise, all the label name is 'noise'
+            # the label is string type and we will convert it to integer type in training
             ret.append([
                 chunk, audio_duration, wav_file, start_sample, end_sample,
                 "noise"
@@ -81,7 +83,7 @@ def generate_csv(wav_files,
         split_chunks (bool): audio split flag
     """
     logger.info(f'Generating csv: {output_file}')
-    header = ["utt_id", "duration", "wav", "start", "stop", "lab_id"]
+    header = ["utt_id", "duration", "wav", "start", "stop", "label"]
     csv_lines = []
     for item in tqdm.tqdm(wav_files):
         csv_lines.extend(

examples/voxceleb/sv0/local/make_vox_csv_dataset_from_json.py

@@ -26,7 +26,7 @@ from yacs.config import CfgNode
 
 from paddleaudio import load as load_audio
 from paddlespeech.s2t.utils.log import Log
-from paddlespeech.vector.utils.utils import get_chunks
+from paddlespeech.vector.utils.vector_utils import get_chunks
 
 logger = Log(__name__).getlog()
 
@@ -38,28 +38,31 @@ def prepare_csv(wav_files, output_file, config, split_chunks=True):
         wav_files (list): all the audio list to prepare the csv file
         output_file (str): the output csv file
         config (CfgNode): yaml configuration content
-        split_chunks (bool): audio split flag
+        split_chunks (bool, optional): audio split flag. Defaults to True.
     """
     if not os.path.exists(os.path.dirname(output_file)):
         os.makedirs(os.path.dirname(output_file))
     csv_lines = []
-    header = ["utt_id", "duration", "wav", "start", "stop", "lab_id"]
+    header = ["utt_id", "duration", "wav", "start", "stop", "label"]
     # voxceleb meta info for each training utterance segment
     # we extract a segment from a utterance to train 
     # and the segment' period is between start and stop time point in the original wav file
-    # each field in the meta means as follows:
-    # utt_id: the utterance segment name
-    # duration: utterance segment time
-    # wav: utterance file path
-    # start: start point in the original wav file
-    # stop: stop point in the original wav file
-    # lab_id: the utterance segment's speaker name
+    # each field in the meta info means as follows:
+    # utt_id: the utterance segment name, which is uniq in training dataset
+    # duration: the total utterance time
+    # wav: utterance file path, which should be absoulute path
+    # start: start point in the original wav file sample point range
+    # stop: stop point in the original wav file sample point range
+    # label: the utterance segment's label name, 
+    #        which is speaker name in speaker verification domain
     for item in tqdm.tqdm(wav_files, total=len(wav_files)):
         item = json.loads(item.strip())
-        audio_id = item['utt'].replace(".wav", "")
+        audio_id = item['utt'].replace(".wav",
+                                       "")  # we remove the wav suffix name
         audio_duration = item['feat_shape'][0]
         wav_file = item['feat']
-        spk_id = audio_id.split('-')[0]
+        label = audio_id.split('-')[
+            0]  # speaker name in speaker verification domain
         waveform, sr = load_audio(wav_file)
         if split_chunks:
             uniq_chunks_list = get_chunks(config.chunk_duration, audio_id,
@@ -68,14 +71,15 @@ def prepare_csv(wav_files, output_file, config, split_chunks=True):
                 s, e = chunk.split("_")[-2:]  # Timestamps of start and end
                 start_sample = int(float(s) * sr)
                 end_sample = int(float(e) * sr)
-                # id, duration, wav, start, stop, spk_id
+                # id, duration, wav, start, stop, label
+                # in vector, the label in speaker id
                 csv_lines.append([
                     chunk, audio_duration, wav_file, start_sample, end_sample,
-                    spk_id
+                    label
                 ])
         else:
             csv_lines.append([
-                audio_id, audio_duration, wav_file, 0, waveform.shape[0], spk_id
+                audio_id, audio_duration, wav_file, 0, waveform.shape[0], label
             ])
 
     with open(output_file, mode="w") as csv_f:
@@ -113,6 +117,9 @@ def get_enroll_test_list(dataset_list, verification_file):
     for dataset in dataset_list:
         with open(dataset, 'r') as f:
             for line in f:
+                # audio_id may be in enroll and test at the same time
+                # eg: 1 a.wav a.wav
+                # the audio a.wav is enroll and test file at the same time
                 audio_id = json.loads(line.strip())['utt']
                 if audio_id in enroll_audios:
                     enroll_files.append(line)
@@ -145,17 +152,18 @@ def get_train_dev_list(dataset_list, target_dir, split_ratio):
     for dataset in dataset_list:
         with open(dataset, 'r') as f:
             for line in f:
-                spk_id = json.loads(line.strip())['utt2spk']
-                speakers.add(spk_id)
+                # the label is speaker name
+                label_name = json.loads(line.strip())['utt2spk']
+                speakers.add(label_name)
                 audio_files.append(line.strip())
     speakers = sorted(speakers)
     logger.info(f"we get {len(speakers)} speakers from all the train dataset")
 
-    with open(os.path.join(target_dir, "meta", "spk_id2label.txt"), 'w') as f:
-        for label, spk_id in enumerate(speakers):
-            f.write(f'{spk_id} {label}\n')
+    with open(os.path.join(target_dir, "meta", "label2id.txt"), 'w') as f:
+        for label_id, label_name in enumerate(speakers):
+            f.write(f'{label_name} {label_id}\n')
     logger.info(
-        f'we store the speakers to {os.path.join(target_dir, "meta", "spk_id2label.txt")}'
+        f'we store the speakers to {os.path.join(target_dir, "meta", "label2id.txt")}'
     )
 
     # the split_ratio is for train dataset 
@@ -185,7 +193,7 @@ def prepare_data(args, config):
         return
 
     # stage 1: prepare the enroll and test csv file
-    #          And we generate the speaker to label file spk_id2label.txt
+    #          And we generate the speaker to label file label2id.txt
     logger.info("start to prepare the data csv file")
     enroll_files, test_files = get_enroll_test_list(
         [args.test], verification_file=config.verification_file)

paddlespeech/vector/exps/ecapa_tdnn/test.py

@@ -21,10 +21,10 @@ from paddle.io import DataLoader
 from tqdm import tqdm
 from yacs.config import CfgNode
 
-from paddleaudio.datasets import VoxCeleb
 from paddleaudio.metric import compute_eer
 from paddlespeech.s2t.utils.log import Log
 from paddlespeech.vector.io.batch import batch_feature_normalize
+from paddlespeech.vector.io.dataset import CSVDataset
 from paddlespeech.vector.models.ecapa_tdnn import EcapaTdnn
 from paddlespeech.vector.modules.sid_model import SpeakerIdetification
 from paddlespeech.vector.training.seeding import seed_everything
@@ -58,9 +58,8 @@ def main(args, config):
 
     # stage4: construct the enroll and test dataloader
 
-    enroll_dataset = VoxCeleb(
-        subset='enroll',
-        target_dir=args.data_dir,
+    enroll_dataset = CSVDataset(
+        os.path.join(args.data_dir, "vox/csv/enroll.csv"),
         feat_type='melspectrogram',
         random_chunk=False,
         n_mels=config.n_mels,
@@ -69,15 +68,14 @@ def main(args, config):
     enroll_sampler = BatchSampler(
         enroll_dataset, batch_size=config.batch_size,
         shuffle=True)  # Shuffle to make embedding normalization more robust.
-    enrol_loader = DataLoader(enroll_dataset,
+    enroll_loader = DataLoader(enroll_dataset,
                     batch_sampler=enroll_sampler,
                     collate_fn=lambda x: batch_feature_normalize(
-                            x, mean_norm=True, std_norm=False),
+                                x, mean_norm=True, std_norm=False),
                     num_workers=config.num_workers,
                     return_list=True,)
-    test_dataset = VoxCeleb(
-        subset='test',
-        target_dir=args.data_dir,
+    test_dataset = CSVDataset(
+        os.path.join(args.data_dir, "vox/csv/test.csv"),
         feat_type='melspectrogram',
         random_chunk=False,
         n_mels=config.n_mels,
@@ -108,9 +106,9 @@ def main(args, config):
     id2embedding = {}
     # Run multi times to make embedding normalization more stable.
     for i in range(2):
-        for dl in [enrol_loader, test_loader]:
+        for dl in [enroll_loader, test_loader]:
             logger.info(
-                f'Loop {[i+1]}: Computing embeddings on {dl.dataset.subset} dataset'
+                f'Loop {[i+1]}: Computing embeddings on {dl.dataset.csv_path} dataset'
             )
             with paddle.no_grad():
                 for batch_idx, batch in enumerate(tqdm(dl)):
@@ -152,8 +150,8 @@ def main(args, config):
     labels = []
     enroll_ids = []
     test_ids = []
-    logger.info(f"read the trial from {VoxCeleb.veri_test_file}")
-    with open(VoxCeleb.veri_test_file, 'r') as f:
+    logger.info(f"read the trial from {config.verification_file}")
+    with open(config.verification_file, 'r') as f:
         for line in f.readlines():
             label, enroll_id, test_id = line.strip().split(' ')
             labels.append(int(label))

paddlespeech/vector/exps/ecapa_tdnn/train.py

@@ -57,12 +57,10 @@ def main(args, config):
     # note: some cmd must do in rank==0, so wo will refactor the data prepare code
     train_dataset = CSVDataset(
         csv_path=os.path.join(args.data_dir, "vox/csv/train.csv"),
-        spk_id2label_path=os.path.join(args.data_dir,
-                                       "vox/meta/spk_id2label.txt"))
+        label2id_path=os.path.join(args.data_dir, "vox/meta/label2id.txt"))
     dev_dataset = CSVDataset(
         csv_path=os.path.join(args.data_dir, "vox/csv/dev.csv"),
-        spk_id2label_path=os.path.join(args.data_dir,
-                                       "vox/meta/spk_id2label.txt"))
+        label2id_path=os.path.join(args.data_dir, "vox/meta/label2id.txt"))
 
     if config.augment:
         augment_pipeline = build_augment_pipeline(target_dir=args.data_dir)
@@ -148,7 +146,6 @@ def main(args, config):
         train_reader_cost = 0.0
         train_feat_cost = 0.0
         train_run_cost = 0.0
-        train_misce_cost = 0.0
 
         reader_start = time.time()
         for batch_idx, batch in enumerate(train_loader):

paddlespeech/vector/io/dataset.py

@@ -25,7 +25,7 @@ logger = Log(__name__).getlog()
 # wav: utterance file path
 # start: start point in the original wav file
 # stop: stop point in the original wav file
-# lab_id: the utterance segment's label id
+# label: the utterance segment's label id
 
 
 @dataclass
@@ -45,24 +45,24 @@ class meta_info:
     wav: str
     start: int
     stop: int
-    lab_id: str
+    label: str
 
 
 class CSVDataset(Dataset):
-    def __init__(self, csv_path, spk_id2label_path=None, config=None):
+    def __init__(self, csv_path, label2id_path=None, config=None):
         """Implement the CSV Dataset
 
         Args:
             csv_path (str): csv dataset file path
-            spk_id2label_path (str): the utterance label to integer id map file path
+            label2id_path (str): the utterance label to integer id map file path
             config (CfgNode): yaml config
         """
         super().__init__()
         self.csv_path = csv_path
-        self.spk_id2label_path = spk_id2label_path
+        self.label2id_path = label2id_path
         self.config = config
-        self.spk_id2label = {}
-        self.label2spk_id = {}
+        self.id2label = {}
+        self.label2id = {}
         self.data = self.load_data_csv()
         self.load_speaker_to_label()
 
@@ -71,7 +71,7 @@ class CSVDataset(Dataset):
         the csv dataset's format has six fields, 
         that is audio_id or utt_id, audio duration, segment start point, segment stop point 
         and utterance label.
-        Note in training period, the utterance label must has a map to integer id in spk_id2label_path 
+        Note in training period, the utterance label must has a map to integer id in label2id_path 
         """
         data = []
 
@@ -91,16 +91,15 @@ class CSVDataset(Dataset):
         The speaker label is real speaker label in speaker verification domain,
         and in language identification is language label.
         """
-        if not self.spk_id2label_path:
+        if not self.label2id_path:
             logger.warning("No speaker id to label file")
             return
-        self.spk_id2label = {}
-        self.label2spk_id = {}
-        with open(self.spk_id2label_path, 'r') as f:
+
+        with open(self.label2id_path, 'r') as f:
             for line in f.readlines():
-                spk_id, label = line.strip().split(' ')
-                self.spk_id2label[spk_id] = int(label)
-                self.label2spk_id[int(label)] = spk_id
+                label_name, label_id = line.strip().split(' ')
+                self.label2id[label_name] = int(label_id)
+                self.id2label[int(label_id)] = label_name
 
     def convert_to_record(self, idx: int):
         """convert the dataset sample to training record the CSV Dataset
@@ -130,8 +129,8 @@ class CSVDataset(Dataset):
         # we only return the waveform as feat
         waveform = waveform[start:stop]
         record.update({'feat': waveform})
-        if self.spk_id2label:
-            record.update({'label': self.spk_id2label[record['lab_id']]})
+        if self.label2id:
+            record.update({'label': self.label2id[record['label']]})
 
         return record
 

paddlespeech/vector/utils/utils.py → paddlespeech/vector/utils/vector_utils.py

@@ -17,14 +17,14 @@ def get_chunks(seg_dur, audio_id, audio_duration):
     """Get all chunk segments from a utterance
 
     Args:
-        seg_dur (float): segment chunk duration
-        audio_id (str): utterance name
-        audio_duration (float): utterance duration
+        seg_dur (float): segment chunk duration, seconds
+        audio_id (str): utterance name, 
+        audio_duration (float): utterance duration, seconds
 
     Returns:
         List: all the chunk segments 
     """
-    num_chunks = int(audio_duration / seg_dur)  # all in milliseconds
+    num_chunks = int(audio_duration / seg_dur)  # all in seconds
     chunk_lst = [
         audio_id + "_" + str(i * seg_dur) + "_" + str(i * seg_dur + seg_dur)
         for i in range(num_chunks)