jsonlines reade manifest file

paddlespeech/s2t/frontend/augmentor/impulse_response.py

@@ -12,9 +12,9 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 """Contains the impulse response augmentation model."""
+import jsonlines
 from paddlespeech.s2t.frontend.audio import AudioSegment
 from paddlespeech.s2t.frontend.augmentor.base import AugmentorBase
-from paddlespeech.s2t.frontend.utility import read_manifest
 
 
 class ImpulseResponseAugmentor(AugmentorBase):
@@ -28,7 +28,8 @@ class ImpulseResponseAugmentor(AugmentorBase):
 
     def __init__(self, rng, impulse_manifest_path):
         self._rng = rng
-        self._impulse_manifest = read_manifest(impulse_manifest_path)
+        with jsonlines.open(impulse_manifest_path, 'r') as reader:
+            self._impulse_manifest = list(reader)
 
     def __call__(self, x, uttid=None, train=True):
         if not train:

paddlespeech/s2t/frontend/augmentor/noise_perturb.py

@@ -12,9 +12,9 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 """Contains the noise perturb augmentation model."""
+import jsonlines
 from paddlespeech.s2t.frontend.audio import AudioSegment
 from paddlespeech.s2t.frontend.augmentor.base import AugmentorBase
-from paddlespeech.s2t.frontend.utility import read_manifest
 
 
 class NoisePerturbAugmentor(AugmentorBase):
@@ -34,7 +34,8 @@ class NoisePerturbAugmentor(AugmentorBase):
         self._min_snr_dB = min_snr_dB
         self._max_snr_dB = max_snr_dB
         self._rng = rng
-        self._noise_manifest = read_manifest(manifest_path=noise_manifest_path)
+        with jsonlines.open(noise_manifest_path, 'r') as reader:
+            self._noise_manifest = list(reader)
 
     def __call__(self, x, uttid=None, train=True):
         if not train:

paddlespeech/s2t/frontend/normalizer.py

@@ -13,10 +13,9 @@
 # limitations under the License.
 """Contains feature normalizers."""
 import json
-
+import jsonlines
 import numpy as np
 import paddle
-import jsonlines
 from paddle.io import DataLoader
 from paddle.io import Dataset
 
@@ -27,24 +26,6 @@ from paddlespeech.s2t.utils.log import Log
 __all__ = ["FeatureNormalizer"]
 
 logger = Log(__name__).getlog()
-
-def read_manifest(manifest_path):
-     """Load and parse manifest file.
- 
-     Args:
-         manifest_path ([type]): Manifest file to load and parse.
-     Raises:
-         IOError: If failed to parse the manifest.
- 
-     Returns:
-         List[dict]: Manifest parsing results.
-     """
- 
-     manifest = []
-     with jsonlines.open(manifest_path, 'r') as reader:
-         for json_data in reader:
-            manifest.append(json_data)
-     return manifest
  
 # https://github.com/PaddlePaddle/Paddle/pull/31481
 class CollateFunc(object):
@@ -78,10 +59,9 @@ class CollateFunc(object):
 class AudioDataset(Dataset):
     def __init__(self, manifest_path, num_samples=-1, rng=None, random_seed=0):
         self._rng = rng if rng else np.random.RandomState(random_seed)
-        manifest = []
+
         with jsonlines.open(manifest_path, 'r') as reader:
-         for json_data in reader:
-            manifest.append(json_data)
+            manifest = list(reader)
         
         if num_samples == -1:
             sampled_manifest = manifest

paddlespeech/s2t/frontend/utility.py

@@ -64,27 +64,8 @@ def load_dict(dict_path: Optional[Text], maskctc=False) -> Optional[List[Text]]:
         char_list.append(MASKCTC)
     return char_list
 
-
-def read_manifest(manifest_path,):
-    """Load and parse manifest file.
-
-    Args:
-        manifest_path ([type]): Manifest file to load and parse.
-
-    Raises:
-        IOError: If failed to parse the manifest.
-
-    Returns:
-        List[dict]: Manifest parsing results.
-    """
-    manifest = []
-    with jsonlines.open(manifest_path, 'r') as reader:
-        for json_data in reader:
-            manifest.append(json_data)
-    return manifest
-
     
-def read_manifest_filter(
+def read_manifest(
         manifest_path,
         max_input_len=float('inf'),
         min_input_len=0.0,

paddlespeech/s2t/io/dataloader.py

@@ -15,11 +15,11 @@ from typing import Any
 from typing import Dict
 from typing import List
 from typing import Text
+import jsonlines
 
 import numpy as np
 from paddle.io import DataLoader
 
-from paddlespeech.s2t.frontend.utility import read_manifest
 from paddlespeech.s2t.io.batchfy import make_batchset
 from paddlespeech.s2t.io.converter import CustomConverter
 from paddlespeech.s2t.io.dataset import TransformDataset
@@ -91,7 +91,9 @@ class BatchDataLoader():
         self.n_iter_processes = n_iter_processes
 
         # read json data
-        self.data_json = read_manifest(json_file)
+        with jsonlines.open(json_file, 'r') as reader:
+            self.data_json = list(reader)
+            
         self.feat_dim, self.vocab_size = feat_dim_and_vocab_size(
             self.data_json, mode='asr')
 

paddlespeech/s2t/io/dataset.py

@@ -14,7 +14,7 @@
 # Modified from espnet(https://github.com/espnet/espnet)
 # Modified from wenet(https://github.com/wenet-e2e/wenet)
 from typing import Optional
-
+import jsonlines
 from paddle.io import Dataset
 from yacs.config import CfgNode
 
@@ -95,7 +95,7 @@ class ManifestDataset(Dataset):
         super().__init__()
 
         # read manifest
-        self._manifest = read_manifest_filter(
+        self._manifest = read_manifest(
             manifest_path=manifest_path,
             max_input_len=max_input_len,
             min_input_len=min_input_len,
@@ -184,7 +184,8 @@ class AudioDataset(Dataset):
         """
         assert batch_type in ['static', 'dynamic']
         # read manifest
-        data = read_manifest(data_file)
+        with jsonlines.open(data_file, 'r') as reader:
+            data = list(reader)
         if sort:
             data = sorted(data, key=lambda x: x["feat_shape"][0])
         if raw_wav:

paddlespeech/s2t/utils/socket_server.py

@@ -20,8 +20,7 @@ import time
 import wave
 from time import gmtime
 from time import strftime
-
-from paddlespeech.s2t.frontend.utility import read_manifest
+import jsonlines
 
 __all__ = ["socket_send", "warm_up_test", "AsrTCPServer", "AsrRequestHandler"]
 
@@ -44,7 +43,8 @@ def warm_up_test(audio_process_handler,
                  num_test_cases,
                  random_seed=0):
     """Warming-up test."""
-    manifest = read_manifest(manifest_path)
+    with jsonlines.open(manifest_path) as reader:
+        manifest = list(reader)
     rng = random.Random(random_seed)
     samples = rng.sample(manifest, num_test_cases)
     for idx, sample in enumerate(samples):

utils/dump_manifest.py

@@ -16,8 +16,7 @@
 import argparse
 from pathlib import Path
 from typing import Union
-
-from paddlespeech.s2t.frontend.utility import read_manifest
+import jsonlines
 
 key_whitelist = set(['feat', 'text', 'syllable', 'phone'])
 filename = {
@@ -32,7 +31,10 @@ def dump_manifest(manifest_path, output_dir: Union[str, Path]):
 
     output_dir = Path(output_dir).expanduser()
     manifest_path = Path(manifest_path).expanduser()
-    manifest_jsons = read_manifest(manifest_path)
+
+    with jsonlines.open(str(manifest_path), 'r') as reader:
+        manifest_jsons = list(reader)
+        
     first_line = manifest_jsons[0]
     file_map = {}
 

utils/format_data.py

@@ -15,11 +15,11 @@
 """format manifest with more metadata."""
 import argparse
 import functools
+import jsonlines
 import json
 
 from paddlespeech.s2t.frontend.featurizer.text_featurizer import TextFeaturizer
 from paddlespeech.s2t.frontend.utility import load_cmvn
-from paddlespeech.s2t.frontend.utility import read_manifest
 from paddlespeech.s2t.io.utility import feat_type
 from paddlespeech.s2t.utils.utility import add_arguments
 from paddlespeech.s2t.utils.utility import print_arguments
@@ -71,7 +71,9 @@ def main():
     # }
     count = 0
     for manifest_path in args.manifest_paths:
-        manifest_jsons = read_manifest(manifest_path)
+        with jsonlines.open(str(manifest_path), 'r') as reader:
+            manifest_jsons = list(reader)
+        
         for line_json in manifest_jsons:
             output_json = {
                 "input": [],

utils/format_triplet_data.py

@@ -16,10 +16,10 @@
 import argparse
 import functools
 import json
+import jsonlines
 
 from paddlespeech.s2t.frontend.featurizer.text_featurizer import TextFeaturizer
 from paddlespeech.s2t.frontend.utility import load_cmvn
-from paddlespeech.s2t.frontend.utility import read_manifest
 from paddlespeech.s2t.io.utility import feat_type
 from paddlespeech.s2t.utils.utility import add_arguments
 from paddlespeech.s2t.utils.utility import print_arguments
@@ -63,7 +63,8 @@ def main():
 
     count = 0
     for manifest_path in args.manifest_paths:
-        manifest_jsons = read_manifest(manifest_path)
+        with jsonlines.open(str(manifest_path), 'r') as reader:
+            manifest_jsons = list(reader)
         for line_json in manifest_jsons:
             # text: translation text, text1: transcript text.
             # Currently only support joint-vocab, will add separate vocabs setting.

utils/manifest_key_value.py

@@ -3,10 +3,10 @@
 import argparse
 import functools
 from pathlib import Path
+import jsonlines
 
 from utils.utility import add_arguments
 from utils.utility import print_arguments
-from utils.utility import read_manifest
 
 
 def main(args):
@@ -19,7 +19,8 @@ def main(args):
     dur_scp = outdir / 'duration'
     text_scp = outdir / 'text'
 
-    manifest_jsons = read_manifest(args.manifest_path)
+    with jsonlines.open(args.manifest_path, 'r') as reader:
+        manifest_jsons = list(reader)
 
     with wav_scp.open('w') as fwav, dur_scp.open('w') as fdur, text_scp.open(
             'w') as ftxt:

utils/utility.py

@@ -22,32 +22,10 @@ from typing import Text
 __all__ = [
     "check_md5sum", "getfile_insensitive", "download_multi", "download",
     "unpack", "unzip", "md5file", "print_arguments", "add_arguments",
-    "read_manifest", "get_commandline_args"
+    "get_commandline_args"
 ]
 
 
-def read_manifest(manifest_path):
-    """Load and parse manifest file.
-    Args:
-        manifest_path ([type]): Manifest file to load and parse.
-
-    Raises:
-        IOError: If failed to parse the manifest.
-
-    Returns:
-        List[dict]: Manifest parsing results.
-    """
-
-    manifest = []
-    for json_line in open(manifest_path, 'r'):
-        try:
-            json_data = json.loads(json_line)
-            manifest.append(json_data)
-        except Exception as e:
-            raise IOError("Error reading manifest: %s" % str(e))
-    return manifest
-
-
 def get_commandline_args():
     extra_chars = [
         " ",