sonusai 0.17.0__py3-none-any.whl → 0.17.3__py3-none-any.whl

sonusai/utils/__init__.py

@@ -27,11 +27,12 @@ from .max_text_width import max_text_width
 from .model_utils import import_module
 from .numeric_conversion import float_to_int16
 from .numeric_conversion import int16_to_float
-from .onnx_utils import SonusAIMetaData
 from .onnx_utils import add_sonusai_metadata
 from .onnx_utils import get_sonusai_metadata
+from .onnx_utils import load_ort_session
 from .parallel import pp_imap
 from .parallel import pp_tqdm_imap
+from .path_info import PathInfo
 from .print_mixture_details import print_class_count
 from .print_mixture_details import print_mixture_details
 from .ranges import consolidate_range
@@ -48,5 +49,5 @@ from .stacked_complex import stacked_complex_imag
 from .stacked_complex import stacked_complex_real
 from .stacked_complex import unstack_complex
 from .stratified_shuffle_split import stratified_shuffle_split_mixid
-from .wave import write_wav
+from .write_audio import write_audio
 from .yes_or_no import yes_or_no

sonusai/utils/asr_functions/aaware_whisper.py

@@ -13,7 +13,7 @@ def aaware_whisper(data: ASRData) -> ASRResult:
     from sonusai import SonusAIError
     from sonusai.utils import ASRResult
     from sonusai.utils import float_to_int16
-    from sonusai.utils import write_wav
+    from sonusai.utils import write_audio
 
     url = getenv('AAWARE_WHISPER_URL')
     if url is None:
@@ -22,7 +22,7 @@ def aaware_whisper(data: ASRData) -> ASRResult:
 
     with tempfile.TemporaryDirectory() as tmp:
         file = join(tmp, 'asr.wav')
-        write_wav(name=file, audio=float_to_int16(data.audio))
+        write_audio(name=file, audio=float_to_int16(data.audio))
 
         files = {'audio_file': (file, open(file, 'rb'), 'audio/wav')}
 

sonusai/utils/asr_manifest_functions/__init__.py

@@ -1,4 +1,3 @@
-from .data import PathInfo
 from .data import TranscriptData
 from .librispeech import collect_librispeech_transcripts
 from .librispeech import get_librispeech_manifest_entry

sonusai/utils/asr_manifest_functions/data.py

@@ -1,9 +1 @@
-from dataclasses import dataclass
-
 TranscriptData = dict[str, str]
-
-
-@dataclass(frozen=True)
-class PathInfo:
-    abs_path: str
-    audio_filepath: str

sonusai/utils/asr_manifest_functions/librispeech.py

@@ -1,4 +1,4 @@
-from sonusai.utils.asr_manifest_functions import PathInfo
+from sonusai.utils import PathInfo
 from sonusai.utils.asr_manifest_functions import TranscriptData
 
 

sonusai/utils/asr_manifest_functions/mcgill_speech.py

@@ -1,4 +1,4 @@
-from sonusai.utils.asr_manifest_functions import PathInfo
+from sonusai.utils import PathInfo
 
 
 def get_mcgill_speech_manifest_entry(entry: PathInfo, transcript_data: list[str]) -> dict:

sonusai/utils/asr_manifest_functions/vctk_noisy_speech.py

@@ -1,4 +1,4 @@
-from sonusai.utils.asr_manifest_functions import PathInfo
+from sonusai.utils import PathInfo
 from sonusai.utils.asr_manifest_functions import TranscriptData
 
 

sonusai/utils/braced_glob.py

@@ -1,9 +1,10 @@
 from typing import Generator
+from typing import LiteralString
 from typing import Optional
 from typing import Set
 
 
-def expand_braces(text: str, seen: Optional[Set[str]] = None) -> Generator[str, None, None]:
+def expand_braces(text: LiteralString | str | bytes, seen: Optional[Set[str]] = None) -> Generator[str, None, None]:
     """Brace-expansion pre-processing for glob.
 
     Expand all the braces, then run glob on each of the results.
@@ -16,6 +17,9 @@ def expand_braces(text: str, seen: Optional[Set[str]] = None) -> Generator[str,
     if seen is None:
         seen = set()
 
+    if not isinstance(text, str):
+        text = str(text)
+
     spans = [m.span() for m in re.finditer(r'\{[^{}]*}', text)][::-1]
     alts = [text[start + 1: stop - 1].split(',') for start, stop in spans]
 
@@ -31,7 +35,7 @@ def expand_braces(text: str, seen: Optional[Set[str]] = None) -> Generator[str,
             yield from expand_braces(''.join(replaced), seen)
 
 
-def braced_glob(pathname: str, recursive: bool = False) -> list[str]:
+def braced_glob(pathname: LiteralString | str | bytes, recursive: bool = False) -> list[str]:
     from glob import glob
 
     result = []
@@ -41,7 +45,7 @@ def braced_glob(pathname: str, recursive: bool = False) -> list[str]:
     return result
 
 
-def braced_iglob(pathname: str, recursive: bool = False) -> Generator[str, None, None]:
+def braced_iglob(pathname: LiteralString | str | bytes, recursive: bool = False) -> Generator[str, None, None]:
     from glob import iglob
 
     for expanded_path in expand_braces(pathname):

sonusai/utils/onnx_utils.py

@@ -1,120 +1,134 @@
-from dataclasses import dataclass
+from typing import Optional
+from typing import Sequence
 
-from sonusai import logger
-from typing import Any  #List, Optional, Tuple
-import onnxruntime as ort
-from onnxruntime import InferenceSession
-import onnx
+from onnx import ModelProto
 from onnx import ValueInfoProto
-from os.path import basename, splitext, exists, isfile
+from onnxruntime import InferenceSession
+from onnxruntime import NodeArg
+from onnxruntime import SessionOptions
+
+REQUIRED_HPARAMS = ('feature', 'batch_size', 'timesteps', 'flatten', 'add1ch', 'truth_mutex')
+
+
+def _extract_shapes(io: list[ValueInfoProto]) -> list[list[int] | str]:
+    shapes: list[list[int] | str] = []
+
+    # iterate through inputs of the graph to find shapes
+    for item in io:
+        # get tensor type: 0, 1, 2, etc.
+        tensor_type = item.type.tensor_type
+        # check if it has a shape
+        if tensor_type.HasField('shape'):
+            tmp_shape = []
+            # iterate through dimensions of the shape
+            for d in tensor_type.shape.dim:
+                if d.HasField('dim_value'):
+                    # known dimension, int value
+                    tmp_shape.append(d.dim_value)
+                elif d.HasField('dim_param'):
+                    # dynamic dim with symbolic name of d.dim_param; set size to 0
+                    tmp_shape.append(0)
+                else:
+                    # unknown dimension with no name; also set to 0
+                    tmp_shape.append(0)
+            # add as a list
+            shapes.append(tmp_shape)
+        else:
+            shapes.append('unknown rank')
 
+    return shapes
 
-@dataclass(frozen=True)
-class SonusAIMetaData:
-    input_shape: list[int]
-    output_shape: list[int]
-    flattened: bool
-    timestep: bool
-    channel: bool
-    mutex: bool
-    feature: str
 
+def get_and_check_inputs(model: ModelProto) -> tuple[list[ValueInfoProto], list[list[int] | str]]:
+    from sonusai import logger
 
-def get_and_check_inputs(model: onnx.ModelProto) -> tuple[list[ValueInfoProto], list[list[int] | str]]:
-    # ignore initializer inputs (only seen in older onnx < v1.5
+    # ignore initializer inputs (only seen in older ONNX < v1.5)
     initializer_names = [x.name for x in model.graph.initializer]
-    onnx_inputs = [ipt for ipt in model.graph.input if ipt.name not in initializer_names]
-    if len(onnx_inputs) != 1:
-        logger.warning(f'Warning: onnx model does not have 1 input, but {len(onnx_inputs)}')
-        #raise SystemExit(1)
-
-    inshapes = []
-    for inp in onnx_inputs:                      # iterate through inputs of the graph to find shapes
-        tensor_type = inp.type.tensor_type       # get tensor type: 0, 1, 2,
-        if (tensor_type.HasField("shape")):      # check if it has a shape:
-            tmpshape = []
-            for d in tensor_type.shape.dim:      # iterate through dimensions of the shape
-                if (d.HasField("dim_value")):    # known dimension, int value
-                    tmpshape.append(d.dim_value)
-                elif (d.HasField("dim_param")):  # dynamic dim with symbolic name of d.dim_param
-                    tmpshape.append(0)           # set size to 0
-                else:                            # unknown dimension with no name
-                    tmpshape.append(0)           # also set to 0
-            inshapes.append(tmpshape)            # add as a list
-        else:
-            inshapes.append("unknown rank")
+    inputs = [i for i in model.graph.input if i.name not in initializer_names]
+    if len(inputs) != 1:
+        logger.warning(f'Warning: ONNX model has {len(inputs)} inputs; expected only 1')
 
     # This one-liner works only if input has type and shape, returns a list
-    #in0shape = [d.dim_value for d in onnx_inputs[0].type.tensor_type.shape.dim]
-
-    return onnx_inputs, inshapes
-
-
-def get_and_check_outputs(model: onnx.ModelProto) -> tuple[list[ValueInfoProto], list[list[int | Any] | str]]:
-    onnx_outputs = [opt for opt in model.graph.output]
-    if len(onnx_outputs) != 1:
-        logger.warning(f'Warning: onnx model does not have 1 output, but {len(onnx_outputs)}')
-
-    oshapes = []
-    for inp in onnx_outputs:                     # iterate through inputs of the graph to find shapes
-        tensor_type = inp.type.tensor_type       # get tensor type: 0, 1, 2,
-        if (tensor_type.HasField("shape")):      # check if it has a shape:
-            tmpshape = []
-            for d in tensor_type.shape.dim:      # iterate through dimensions of the shape
-                if (d.HasField("dim_value")):    # known dimension, int value
-                    tmpshape.append(d.dim_value)
-                elif (d.HasField("dim_param")):  # dynamic dim with symbolic name of d.dim_param
-                    tmpshape.append(0)           # set size to 0
-                else:                            # unknown dimension with no name
-                    tmpshape.append(0)           # also set to 0
-            oshapes.append(tmpshape)             # add as a list
-        else:
-            oshapes.append("unknown rank")
+    # shape0 = [d.dim_value for d in inputs[0].type.tensor_type.shape.dim]
+    shapes = _extract_shapes(inputs)
+
+    return inputs, shapes
+
+
+def get_and_check_outputs(model: ModelProto) -> tuple[list[ValueInfoProto], list[list[int] | str]]:
+    from sonusai import logger
+
+    outputs = [o for o in model.graph.output]
+    if len(outputs) != 1:
+        logger.warning(f'Warning: ONNX model has {len(outputs)} outputs; expected only 1')
 
-    return onnx_outputs, oshapes
+    shapes = _extract_shapes(outputs)
 
+    return outputs, shapes
 
-def add_sonusai_metadata(model, hparams):
-    """Add SonusAI hyper-parameter metadata to an ONNX model using key hparams
+
+def add_sonusai_metadata(model: ModelProto, hparams: dict) -> ModelProto:
+    """Add SonusAI hyperparameters as metadata to an ONNX model using 'hparams' key
 
     :param model: ONNX model
-    :hparams: dictionary of hyper-parameters, added
+    :param hparams: dictionary of hyperparameters to add
+    :return: ONNX model
+
     Note SonusAI conventions require models to have:
-      - feature: Model feature type
-      - is_flattened: Model input feature data is flattened (stride + bins combined)
-      - timesteps: Size of timestep dimension (0 for no dimension)
-      - add1ch:   Model input has channel dimension
-      - truth_mutex: Model label output is mutually exclusive
+        feature: Model feature type
+        batch_size: Model batch size
+        timesteps: Size of timestep dimension (0 for no dimension)
+        flatten: Model input feature data is flattened (stride + bins combined)
+        add1ch: Model input has channel dimension
+        truth_mutex: Model label output is mutually exclusive
     """
+    from sonusai import logger
+
+    # Note hparams should be a dict (i.e., extracted from checkpoint)
+    assert eval(str(hparams)) == hparams
+    for key in REQUIRED_HPARAMS:
+        if key not in hparams.keys():
+            logger.warning(f'Warning: SonusAI hyperparameters are missing: {key}')
 
-    # Add hyper-parameters as metadata in onnx model under hparams key
-    assert eval(str(hparams)) == hparams       # Note hparams should be a dict (i.e. extracted from checkpoint)
     meta = model.metadata_props.add()
-    meta.key = "hparams"
+    meta.key = 'hparams'
     meta.value = str(hparams)
 
     return model
 
 
-def get_sonusai_metadata(session: InferenceSession) -> SonusAIMetaData:
-    """Get SonusAI hyper-parameter metadata from an ONNX Runtime session.
-       Returns dictionary hparams
+def get_sonusai_metadata(session: InferenceSession) -> Optional[dict]:
+    """Get SonusAI hyperparameter metadata from an ONNX Runtime session.
     """
+    from sonusai import logger
+
     meta = session.get_modelmeta()
-    hparams = eval(meta.custom_metadata_map["hparams"])
+    if 'hparams' not in meta.custom_metadata_map.keys():
+        logger.warning("Warning: ONNX model metadata does not contain 'hparams'")
+        return None
 
-    # m = model.get_modelmeta().custom_metadata_map
-    # return SonusAIMetaData(input_shape=model.get_inputs()[0].shape,
-    #                        output_shape=model.get_outputs()[0].shape,
-    #                        flattened=m['is_flattened'] == 'True',
-    #                        timestep=m['has_timestep'] == 'True',
-    #                        channel=m['has_channel'] == 'True',
-    #                        mutex=m['is_mutex'] == 'True',
-    #                        feature=m['feature'])
+    hparams = eval(meta.custom_metadata_map['hparams'])
+    for key in REQUIRED_HPARAMS:
+        if key not in hparams.keys():
+            logger.warning(f'Warning: ONNX model does not have required SonusAI hyperparameters: {key}')
 
     return hparams
 
-def load_ort_session(model_path, providers=['CPUExecutionProvider']):
+
+def load_ort_session(model_path: str, providers: Sequence[str | tuple[str, dict]] = None) -> tuple[
+    InferenceSession, SessionOptions, str, dict, list[NodeArg], list[NodeArg]]:
+    from os.path import basename
+    from os.path import exists
+    from os.path import isfile
+    from os.path import splitext
+
+    import onnxruntime as ort
+
+    from sonusai import logger
+
+    if providers is None:
+        providers = ['CPUExecutionProvider']
+
     if exists(model_path) and isfile(model_path):
         model_basename = basename(model_path)
         model_root = splitext(model_basename)[0]
@@ -123,32 +137,22 @@ def load_ort_session(model_path, providers=['CPUExecutionProvider']):
             session = ort.InferenceSession(model_path, providers=providers)
             options = ort.SessionOptions()
         except Exception as e:
-            logger.exception(f'Error: could not load onnx model from {model_path}: {e}')
+            logger.exception(f'Error: could not load ONNX model from {model_path}: {e}')
             raise SystemExit(1)
     else:
         logger.exception(f'Error: model file does not exist: {model_path}')
         raise SystemExit(1)
 
     logger.info(f'Opened session with provider options: {session._provider_options}.')
-    try:
-        meta = session.get_modelmeta()
-        hparams = eval(meta.custom_metadata_map["hparams"])
-        logger.info(f'Sonusai hyper-parameter metadata was found in model with {len(hparams)} parameters, '
-                    f'checking for required ones ...')
-        # Print to log here will fail if required parameters not available.
-        logger.info(f'feature         {hparams["feature"]}')
-        logger.info(f'batch_size      {hparams["batch_size"]}')
-        logger.info(f'timesteps       {hparams["timesteps"]}')
-        logger.info(f'flatten, add1ch {hparams["flatten"]}, {hparams["add1ch"]}')
-        logger.info(f'truth_mutex     {hparams["truth_mutex"]}')
-    except:
-        hparams = None
-        logger.warning(f'Warning: onnx model does not have required SonusAI hyper-parameters.')
+    hparams = get_sonusai_metadata(session)
+    if hparams is not None:
+        for key in REQUIRED_HPARAMS:
+            logger.info(f'  {key:12} {hparams[key]}')
 
     inputs = session.get_inputs()
     outputs = session.get_outputs()
 
-    #in_names = [n.name for n in session.get_inputs()]
-    #out_names = [n.name for n in session.get_outputs()]
+    # in_names = [n.name for n in session.get_inputs()]
+    # out_names = [n.name for n in session.get_outputs()]
 
-    return session, options, model_root, hparams, inputs, outputs
+    return session, options, model_root, hparams, inputs, outputs

sonusai/utils/path_info.py

@@ -0,0 +1,7 @@
+from dataclasses import dataclass
+
+
+@dataclass(frozen=True)
+class PathInfo:
+    abs_path: str
+    audio_filepath: str

sonusai/utils/{wave.py → write_audio.py}

@@ -2,8 +2,8 @@ from sonusai.mixture.constants import SAMPLE_RATE
 from sonusai.mixture.datatypes import AudioT
 
 
-def write_wav(name: str, audio: AudioT, sample_rate: int = SAMPLE_RATE) -> None:
-    """ Write a simple, uncompressed WAV file.
+def write_audio(name: str, audio: AudioT, sample_rate: int = SAMPLE_RATE) -> None:
+    """ Write an audio file.
 
     To write multiple channels, use a 2D array of shape [channels, samples].
     The bits per sample and PCM/float are determined by the data type.

{sonusai-0.17.0.dist-info → sonusai-0.17.3.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sonusai
-Version: 0.17.0
+Version: 0.17.3
 Summary: Framework for building deep neural network models for sound, speech, and voice AI
 Home-page: https://aaware.com
 License: GPL-3.0-only
@@ -21,11 +21,13 @@ Requires-Dist: h5py (>=3.11.0,<4.0.0)
 Requires-Dist: jiwer (>=3.0.3,<4.0.0)
 Requires-Dist: librosa (>=0.10.1,<0.11.0)
 Requires-Dist: matplotlib (>=3.8.0,<4.0.0)
+Requires-Dist: numpy (>=1.26.4,<2.0.0)
 Requires-Dist: onnx (>=1.14.1,<2.0.0)
 Requires-Dist: onnxruntime (>=1.16.1,<2.0.0)
 Requires-Dist: paho-mqtt (>=2.0.0,<3.0.0)
 Requires-Dist: pandas (>=2.1.1,<3.0.0)
 Requires-Dist: pesq (>=0.0.4,<0.0.5)
+Requires-Dist: praatio (>=6.2.0,<7.0.0)
 Requires-Dist: pyaaware (>=1.5.7,<2.0.0)
 Requires-Dist: pyaudio (>=0.2.14,<0.3.0)
 Requires-Dist: pydub (>=0.25.1,<0.26.0)

{sonusai-0.17.0.dist-info → sonusai-0.17.3.dist-info}/RECORD

@@ -1,14 +1,13 @@
 sonusai/__init__.py,sha256=vzTFfRB-NeO-Sm3puySDJOybk3ND_Oj6w0EejQPmH1U,2978
 sonusai/aawscd_probwrite.py,sha256=GukR5owp_0A3DrqSl9fHWULYgclNft4D5OkHIwfxxkc,3698
-sonusai/audiofe.py,sha256=yPtbxeRAzlnPcRESXKNVexvIm6fM4WVxEHjd0w9n5O0,12455
-sonusai/calc_metric_spenh-save.py,sha256=-LR5BtAnYNYKav1B2ZsB7gGevidCsQ91yFEaH8Ycyr8,61765
-sonusai/calc_metric_spenh.py,sha256=_92RWCyxAf7_S61L5oX6o4GuvuOCDKn7BPg7pB0r1kY,61836
+sonusai/audiofe.py,sha256=zOySiYs5ZZm60eMbA7RjhG6C0Ouhaii3WfL1d0Q8rxg,11154
+sonusai/calc_metric_spenh.py,sha256=0UvJibVjjoHTty3QFh0qzugXe-LvIj2XsMwoaIR44BM,62171
 sonusai/data/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 sonusai/data/genmixdb.yml,sha256=-XSs_hUR6wHJVoTPmSewzXL7u61X-xmHY46lNPatxSE,1025
 sonusai/data/speech_ma01_01.wav,sha256=PK0vMKg-NR6rPE3KouxHGF6PKXnJCr7AwjMqfu98LUA,76644
 sonusai/data/whitenoise.wav,sha256=I2umov0m34y56F9IsIBi1XtE76ZeZaSKDf70cJRe3pI,1920044
 sonusai/doc/__init__.py,sha256=rP5Hgn0Iys_xkuv4caxngdqehuU4zLZsiKuv8Nde67M,19
-sonusai/doc/doc.py,sha256=3z210v6ZckuOlsGZ3ySQBdlCNmBp2M1ahqhqG_eUN58,22664
+sonusai/doc/doc.py,sha256=4NEZ2K-hTk7Y1Gxx09UEjNhiYkD9xid-kJ1Nt8H5_gM,22670
 sonusai/doc.py,sha256=l8CaFgLI8mqx4tn0aXfxKqa2dy9GgC0zjYxZAkpmi1E,878
 sonusai/genft.py,sha256=OzET3iTE-QhrUckzidfZvCDXZlAxIF5Xe5NEf856Vvk,5662
 sonusai/genmix.py,sha256=TU5aTebGHsbfwsRbynYbegGBelSma9khuQkDk0dFE3I,7075
@@ -29,28 +28,29 @@ sonusai/metrics/class_summary.py,sha256=4Mb25nuk6eqotnQSFMuOQL3zofGcpNXDfDlPa513
 sonusai/metrics/confusion_matrix_summary.py,sha256=3qg6TMKjJeHtNjj2YnNjPFSlMrQXt0Zcu1dLkGB_aPU,4001
 sonusai/metrics/one_hot.py,sha256=QSeH_GdqBpOAKLrNnQ8gjcPC-vSdUqC0yPEQueTA6VI,13548
 sonusai/metrics/snr_summary.py,sha256=P4U5_Xr7v9F8kF-rZBnpsVNt3p42rIVS6zmch8yfVfg,5575
-sonusai/mixture/__init__.py,sha256=BfSJL91URq8-JDlbtpc5SQoQRWEUXvxKozbuquX4Mok,5326
-sonusai/mixture/audio.py,sha256=S-ZROf5rVvwv1TCEuwJHz1FfX4oVubb4QhbybUMMqtM,2150
+sonusai/mixture/__init__.py,sha256=yszEbRnlxeZXSegEBUVwyrSZwNIl6ufaJu_NiZ-1rqY,5399
+sonusai/mixture/audio.py,sha256=2lqy0DtTMTYhX4aAOIvVtLNn6QB5ivTa7cJIaAlbfAg,2385
 sonusai/mixture/augmentation.py,sha256=Blb90tdTwBOj5w9tRcYyS5H67YJuFiXsGqwZWd7ON4g,10468
 sonusai/mixture/class_count.py,sha256=_wFnVl2yEOnbor7pLg7cYOUeX6nioov-03Cv3SEbh2k,996
 sonusai/mixture/config.py,sha256=d2IzZ1samHWGMpoKzSmUwMyAWWhgmyNoxyO8oiUwbsg,22193
 sonusai/mixture/constants.py,sha256=xjCskcQi6khqYZDf7j6z1OkeN1C6wE06kBBapcJiNI4,1428
-sonusai/mixture/datatypes.py,sha256=zaxfOHw8ddt-i8JPYOPnlqWz_EHBEDoO4q2VAqJViHM,8173
+sonusai/mixture/datatypes.py,sha256=mMNxtzyDvAmtuoTHVVJP7jBi6OH-QyC1NfC_ZIiuLlY,8440
 sonusai/mixture/eq_rule_is_valid.py,sha256=MpQwRA5M76wSiQWEI1lW2cLFdPaMttBLcQp3tWD8efM,1243
 sonusai/mixture/feature.py,sha256=Rwuf82IoXzhHPGbKYVGcatImF_ssBf_FfvbqghVPXtg,4116
 sonusai/mixture/generation.py,sha256=miUrc3QOSUNIG6mDkiMCZ6M2ulivUZxlYUAJUOVomWc,39039
 sonusai/mixture/helpers.py,sha256=GSGSD2KnvOeEIB6IwNTxyaQNjghTSBMB729kUEd_RiM,22403
 sonusai/mixture/log_duration_and_sizes.py,sha256=baTUpqyM15wA125jo9E3posmVJUe3WlpksyO6v9Jul0,1347
 sonusai/mixture/mapped_snr_f.py,sha256=mlbYM1t14OXe_Zg4CjpWTuA_Zun4W0O3bSUXeodRBQs,1845
-sonusai/mixture/mixdb.py,sha256=9Pe0mEG8pnEf9NZynTIldc05GfdOrgmcVoIt63RG5DA,45279
-sonusai/mixture/soundfile_audio.py,sha256=Ow_IWIMz4pMsLxMP_JsQ8AuHLCWlYQinLa58CFW97f8,2804
+sonusai/mixture/mixdb.py,sha256=zfBaVOuMKHTDzmwR4hWRpGxFwnV_fJkwLDmn9M1SDds,49791
+sonusai/mixture/soundfile_audio.py,sha256=mHa5SIXsu_uE0j3DO52GydRJrvWSzU_nII-7YJfQ6Qo,4154
 sonusai/mixture/sox_audio.py,sha256=HT3kYA9TP5QPCuoOJdUMnGVN-qY6q96DGL8zxuog76o,12277
-sonusai/mixture/sox_augmentation.py,sha256=F9tBdNvX2guCn7gRppAFrxRnBtjw9q6qAq2_v_A4hh0,4490
+sonusai/mixture/sox_augmentation.py,sha256=kBWPrsFk0EBi71nLcKt5v0GA34bY7g9D9x0cEamNWbU,4564
+sonusai/mixture/speaker_metadata.py,sha256=l98avdxLYUsSDZ88xUjfvHnACkbnD0_Dg1aBGDbzS9I,1380
 sonusai/mixture/spectral_mask.py,sha256=8AkCwhy-PSdP1Uri9miKZP-bXFYnFcH_c9xZCGrHavU,2071
 sonusai/mixture/target_class_balancing.py,sha256=NTNiKZH0_PWLooeow0l41CjJKK8ZTMVbUqz9ZkaNtWk,4900
 sonusai/mixture/targets.py,sha256=wyy5vhLhuN-hqBMBGoziVvEJg3FKFvJFgmEE7_LaV2M,7908
 sonusai/mixture/tokenized_shell_vars.py,sha256=gCxw8SQUcal6mqWKF7hOBTgSQmbJUk1nT0Gn3H8GA0U,4705
-sonusai/mixture/torchaudio_audio.py,sha256=qeYlW2G1q47Dml_Lf12Y9L5C0evo-heGw1NT470ZhGo,2395
+sonusai/mixture/torchaudio_audio.py,sha256=KhHeOMsjmbwOaAcoKD61aFvYBYSlA8OysfT5iGn45MA,3010
 sonusai/mixture/torchaudio_augmentation.py,sha256=1vEDHI0caL1vrgoY2lAWe4CiHE2jKRuKKH7x23GHw0w,4390
 sonusai/mixture/truth.py,sha256=Y41pZ52Xkols9LUler0NlgnilUOscBIucmw4GcxXNzU,1612
 sonusai/mixture/truth_functions/__init__.py,sha256=82lKYHhLy8KW3gHngrocoqwupGVLVsWdIXdYs3vhjOc,359
@@ -61,32 +61,36 @@ sonusai/mixture/truth_functions/file.py,sha256=jOJuC_3y9BH6GGOp9eKcbVrHLVRzUA80B
 sonusai/mixture/truth_functions/phoneme.py,sha256=stYdlPuNytQK_LLT61OJLfYSqKd-sDjQZdtJKGzt5wA,479
 sonusai/mixture/truth_functions/sed.py,sha256=8cHjEFjZaH_0hIOHhPmj4AJz2GpEADM6Ys2x4NoiWSY,2469
 sonusai/mixture/truth_functions/target.py,sha256=KAsjugDRooOA5BRcHVAbZRgV7l8S5CFg7CZ0XtKZaQ0,5764
-sonusai/mkmanifest.py,sha256=7lfK7YOdgAEP_Lxrf-YDxZ5iLH9MJuaOltBVpav2M9M,8705
-sonusai/mkwav.py,sha256=kLfC2ZuF-t8P97nqYw2falTZpymxAeXv0YTJCe6nK10,5356
-sonusai/onnx_predict-old.py,sha256=Bz_pR28oAZBarNajlKwyzBxmW7ktum77SmxDN2onKPM,9060
-sonusai/onnx_predict-save.py,sha256=ewiV5-HcW5zcDWuIF9xEbdBwbdL8vNWfu2_kaur5jAo,22354
-sonusai/onnx_predict.py,sha256=-ETvGH7fHXjnmY-c2r_8gOHEX1VSLwdSToZ6gSBP3_w,23021
-sonusai/ovino_predict.py,sha256=QtWY_YEdaeqJL5yikfeeDwhbBtjXYjGNtR9310PGGSc,21830
-sonusai/ovino_query_devices.py,sha256=XkXdOlZldI0MfG6nXZfTq8OgECaY8gAzQl36sJRuaIU,1584
+sonusai/mkmanifest.py,sha256=imI8swwPYVzumrUYEL-9JLvun-ez98PtlUBj2b729k8,8682
+sonusai/mkwav.py,sha256=zfSyIiQTIK3KV9Ij33jkLhhZIMVYqaROcRQ4S7c4sIo,5364
+sonusai/onnx_predict.py,sha256=jSxhD2oFyGSTHOGCXbW4fRT-k4SqKOboK2JaDO-yWcs,8737
 sonusai/plot.py,sha256=ERkmxMM3qjcCDm4LGDQY4fRAncCYAzP7uW8iZ7_brcg,17105
-sonusai/post_spenh_targetf.py,sha256=xOz5T6WZuyTHmfbtILIY9skgH064Wvi2GF2Bo5L3YMU,4998
+sonusai/post_spenh_targetf.py,sha256=pHaJZtms7aj4r6sgqQnEGVi6Gg8H_V29szigogV1vZ8,5002
 sonusai/queries/__init__.py,sha256=oKY5JeqZ4Cz7DwCwPc1_ydB8bUs6KaMcWFp_w02TjOs,255
-sonusai/queries/queries.py,sha256=FNMUKnoY_Ya9S5sNhsB8ppwy0B7V55ilbbjhQRv_UN8,7552
+sonusai/queries/queries.py,sha256=oV-m9uiLZOwYTK-Wo7Gf8dpGisaoGf6uDsAJAarVqZI,7553
+sonusai/speech/__init__.py,sha256=SuPcU_K9wQISsZRIzsRNLtEC6cb616l-Jlx3PU-HWMs,113
+sonusai/speech/l2arctic.py,sha256=28TT3CohvPu98YNUb8O7rWHAYgPGwYTOLSdfNQjOuyc,3736
+sonusai/speech/librispeech.py,sha256=A0IpamojCPXyJiHcjCtI7yNWdMjB00cbggjHslssrg8,3120
+sonusai/speech/mcgill.py,sha256=jcddj64fLdV3sO6CJNafm3w-2SnYoyQtU90odXhdaaE,1976
+sonusai/speech/textgrid.py,sha256=8hB6SdEEXxo6JXVFq8mJ1-ilRbBiRXhaHTQjA-HWg-0,3385
+sonusai/speech/timit.py,sha256=1vWgj6isD3ATOjMJSTjOPLmDkYyB65M5MwYipEmLEvg,4081
+sonusai/speech/types.py,sha256=4eKVPAktpkIrZ2qoVp2iT45zxTVNocQEGT6O_Zlub_w,214
+sonusai/speech/vctk.py,sha256=EAMEBAzjZUI6dw15n-yI2oCN-H4tzM9t4aUVlOxpAbo,1540
+sonusai/speech/voxceleb.py,sha256=aJGN0yDb2LFLmCKmRzmUEjpZWQ-QGWw6XWOpy9967AI,2686
 sonusai/summarize_metric_spenh.py,sha256=OiZe_bhCq5esXNhsOkHDD7g4ssYrpENDHvDVoPzV9iw,1822
-sonusai/torchl_onnx-old.py,sha256=5JYow3XpBaUdtuyAW0mOZyCKL_4FrHvEekYBRdDT6KA,8967
 sonusai/tplot.py,sha256=85T6OPZfxVegHBiSuilFpdgCNMEE0VKAuciNy4rCY5Y,14544
-sonusai/utils/__init__.py,sha256=TCXlcW8W0Up2f5ciSgz3DabvH1MxrrWD0LK6pQTJkeA,2215
+sonusai/utils/__init__.py,sha256=h7QrOyEBMUMoIBFKZpNwDG8Jg-1uw3bs-qflB3CXxhU,2257
 sonusai/utils/asl_p56.py,sha256=-bvQpd-jRQVURbkZJpRoyEAq6gTv9Rc3oFDbh5_lcjY,3861
 sonusai/utils/asr.py,sha256=6y6VYJizHpuQ3MgKbEQ4t2gofO-MW6Ez23oAd6d23IE,2920
 sonusai/utils/asr_functions/__init__.py,sha256=JyHK67s97bw7QzrlkboWhws4yNytdPatqzLJxfwx-yw,43
-sonusai/utils/asr_functions/aaware_whisper.py,sha256=LzO9CZV0wBWkjmCR2nSWN_AW9UJwriAsC1OYSlfVeT8,1981
-sonusai/utils/asr_manifest_functions/__init__.py,sha256=V-w4R7SHUyoeDuMR3tS12j6DGhmfTlUibMKgq1c6ga0,370
-sonusai/utils/asr_manifest_functions/data.py,sha256=mJsaHccBReguOJu9qsshRhL-3GbeyqM0-PXMseFnZbE,151
-sonusai/utils/asr_manifest_functions/librispeech.py,sha256=HIaytcYmjRUkuR6fCQlv3Jh3IDWSox_A6WFcFFAHN9M,1635
-sonusai/utils/asr_manifest_functions/mcgill_speech.py,sha256=2uF9qgBwcue9W9dlRo16Kvr3jX5g53AGjW2rwob-Cqk,957
-sonusai/utils/asr_manifest_functions/vctk_noisy_speech.py,sha256=-69lM0dz18KbU5_-dmSeqDoNNwgJj4UlxgGkNBEi3wM,2169
+sonusai/utils/asr_functions/aaware_whisper.py,sha256=Ew3zb8OfbxEW7q1s-KA7D5eph4SjVSUAJgiLK-vVqhI,1985
+sonusai/utils/asr_manifest_functions/__init__.py,sha256=jfi9xC5c86F_aMSsI5Xj-pxWGxuQ7fwZ8Wdf4T7kDsA,343
+sonusai/utils/asr_manifest_functions/data.py,sha256=nO4oT3EQmydwn1pzc-ZM09yz4X2ic-LQuHzGEnJhKe8,32
+sonusai/utils/asr_manifest_functions/librispeech.py,sha256=_3tGc8qfAUpYJZ0_avpW0vGp7zjdpeqj1HAgXi3TL4Q,1612
+sonusai/utils/asr_manifest_functions/mcgill_speech.py,sha256=dW-5XTC5xOY3PHU2DvlWNWDeoprXDD0Zq2dXDdPAjzE,934
+sonusai/utils/asr_manifest_functions/vctk_noisy_speech.py,sha256=9iMrnE-qabLMnyewyxsBMl0uCS8yS7BPJOdmUoOnGAc,2146
 sonusai/utils/audio_devices.py,sha256=LgaXTln1oRArBzaet3rZiIO2plgtaThuGBc3sJ_sLlo,1414
-sonusai/utils/braced_glob.py,sha256=h4hab7YDbM4CjLg9iSzyHZrkd22IPUOY5zZqHdifkh8,1510
+sonusai/utils/braced_glob.py,sha256=Z_XIpPK17QiP1JbzAnUC5w3oyG8ZovoyM22Wh-Q_vWU,1675
 sonusai/utils/calculate_input_shape.py,sha256=63ILxibYKuTQozY83QN8Y2OOhBEbW_1X47Q0askcHDM,984
 sonusai/utils/convert_string_to_number.py,sha256=i17yIxurp8Iz6NPE-imTRlARrXWqadwm8qbOTuzHZvE,236
 sonusai/utils/create_timestamp.py,sha256=TxoQXWZ3SFdBEHLOv-ujeIsTEJuiFnKOGRy-FQq45YU,148
@@ -103,8 +107,9 @@ sonusai/utils/human_readable_size.py,sha256=SjYT0fUlpbfCzCXHo6csir-VMwqfs5ogr-fg
 sonusai/utils/max_text_width.py,sha256=pxiJMwb_zlkNntexgo7S6lAuF7NLLZvFdOCkxdsQJVY,315
 sonusai/utils/model_utils.py,sha256=lt2KOGJqsinG71W0i3U29UXFO-47GMAlEabsf2um7bA,862
 sonusai/utils/numeric_conversion.py,sha256=GRO_2Fba8CcxcFY7bEXKOEUEUX6neA-VN__Bxi1ULsE,340
-sonusai/utils/onnx_utils.py,sha256=L0BcwF0or1UwxYWzvWNWNJHKvG_oEmI2AxYh4msp2vc,6862
+sonusai/utils/onnx_utils.py,sha256=nh2dUDeuERto-0NnTwZ3a6YKKcZFbZjqLLBVzN2l0IU,5682
 sonusai/utils/parallel.py,sha256=bxedjCzBv9oxzU7NajRr6mOKmkCWr2P7FWAI0p2p9N8,1981
+sonusai/utils/path_info.py,sha256=QY7iQ0nYpeEDnPN9RyPh4DsgYmVYsLrrlAzKuzkqX1o,118
 sonusai/utils/print_mixture_details.py,sha256=BzYM4-wHHNa6zxPzBMUJxwKt0gKHmvbwdd7Yp0w15Yk,3017
 sonusai/utils/ranges.py,sha256=NPBZOVzMb95GTOIxltVO-wSzgcXqZ14wbdV46JDLKrw,1222
 sonusai/utils/read_mixture_data.py,sha256=Sb30RgSpw6DnH_iD81O7G_KOsdfjQWWLk3euEkxfMa8,453
@@ -113,10 +118,10 @@ sonusai/utils/reshape.py,sha256=E8Eu6grynaeWwVO6peIR0BF22SrVaJSa1Rkl109lq6Y,5997
 sonusai/utils/seconds_to_hms.py,sha256=oxLuZhTJJr9swj-fOSOrZJ5vBNM7_BrOMQhX1pYpiv0,260
 sonusai/utils/stacked_complex.py,sha256=feLhz3GC1ILxBGMHOj3sJK--sidsXKbfwkalwAVwizc,2950
 sonusai/utils/stratified_shuffle_split.py,sha256=rJNXvBp-GxoKzH3OpL7k0ANSu5xMP2zJ7K1fm_33UzE,7022
-sonusai/utils/wave.py,sha256=O4ZXkZ6wjrKGa99wBCdFd8G6bp91MXXDnmGihpaEMh0,856
+sonusai/utils/write_audio.py,sha256=ZsPGExwM86QHLLN2LOWekK2uAqf5pV_1oRW811p0QAI,840
 sonusai/utils/yes_or_no.py,sha256=eMLXBVH0cEahiXY4W2KNORmwNQ-ba10eRtldh0y4NYg,263
 sonusai/vars.py,sha256=m2AefF0m5bXWGXpJj8Pi42zWL2ydeEj7bkak3GrtMyM,940
-sonusai-0.17.0.dist-info/METADATA,sha256=wn6MaT5JXlVzp45_huLIAnJeH04kQZm8r0_vWiDX3LU,2443
-sonusai-0.17.0.dist-info/WHEEL,sha256=FMvqSimYX_P7y0a7UY-_Mc83r5zkBZsCYPm7Lr0Bsq4,88
-sonusai-0.17.0.dist-info/entry_points.txt,sha256=zMNjEphEPO6B3cD1GNpit7z-yA9tUU5-j3W2v-UWstU,92
-sonusai-0.17.0.dist-info/RECORD,,
+sonusai-0.17.3.dist-info/METADATA,sha256=Zu3DK9fCT1xiipWOBY4JhUZJ3qm1cVRc5eLAXonDb2s,2522
+sonusai-0.17.3.dist-info/WHEEL,sha256=sP946D7jFCHeNz5Iq4fL4Lu-PrWrFsgfLXbbkciIZwg,88
+sonusai-0.17.3.dist-info/entry_points.txt,sha256=zMNjEphEPO6B3cD1GNpit7z-yA9tUU5-j3W2v-UWstU,92
+sonusai-0.17.3.dist-info/RECORD,,

{sonusai-0.17.0.dist-info → sonusai-0.17.3.dist-info}/WHEEL

@@ -1,4 +1,4 @@
 Wheel-Version: 1.0
-Generator: poetry-core 1.8.1
+Generator: poetry-core 1.9.0
 Root-Is-Purelib: true
 Tag: py3-none-any