sonusai 0.18.9__py3-none-any.whl → 0.19.6__py3-none-any.whl

sonusai/summarize_metric_spenh.py

@@ -13,6 +13,7 @@ Inputs:
     LOC         A SonusAI calc_metric_spenh results directory.
 
 """
+
 import signal
 
 
@@ -21,24 +22,24 @@ def signal_handler(_sig, _frame):
 
     from sonusai import logger
 
-    logger.info('Canceled due to keyboard interrupt')
+    logger.info("Canceled due to keyboard interrupt")
     sys.exit(1)
 
 
 signal.signal(signal.SIGINT, signal_handler)
 
 
-def summarize_metric_spenh(location: str, by: str = 'MIXID', reverse: bool = False) -> str:
+def summarize_metric_spenh(location: str, by: str = "MIXID", reverse: bool = False) -> str:
     import glob
 
     import pandas as pd
 
-    files = sorted(glob.glob(location + '/*_metric_spenh.txt'))
+    files = sorted(glob.glob(location + "/*_metric_spenh.txt"))
     need_header = True
-    header = ['MIXID']
+    header = ["MIXID"]
     data = []
     for file in files:
-        with open(file, 'r') as f:
+        with open(file) as f:
             for i, line in enumerate(f):
                 if i == 1 and need_header:
                     need_header = False
@@ -48,7 +49,7 @@ def summarize_metric_spenh(location: str, by: str = 'MIXID', reverse: bool = Fal
                     break
 
     df = pd.DataFrame(data, columns=header)
-    df[header[0:-2]] = df[header[0:-2]].apply(pd.to_numeric, errors='coerce')
+    df[header[0:-2]] = df[header[0:-2]].apply(pd.to_numeric, errors="coerce")
     return df.sort_values(by=by, ascending=not reverse).to_string(index=False)
 
 
@@ -60,12 +61,12 @@ def main():
 
     args = docopt(trim_docstring(__doc__), version=sonusai.__version__, options_first=True)
 
-    by = args['--sort']
-    reverse = args['--reverse']
-    location = args['LOC']
+    by = args["--sort"]
+    reverse = args["--reverse"]
+    location = args["LOC"]
 
     print(summarize_metric_spenh(location, by, reverse))
 
 
-if __name__ == '__main__':
+if __name__ == "__main__":
     main()

sonusai/utils/__init__.py

@@ -1,8 +1,10 @@
 # SonusAI general utilities
+# ruff: noqa: F401
 from .asl_p56 import asl_p56
 from .asr import ASRResult
 from .asr import calc_asr
 from .asr import get_available_engines
+from .asr import validate_asr
 from .audio_devices import get_default_input_device
 from .audio_devices import get_input_device_index_by_name
 from .audio_devices import get_input_devices
@@ -32,14 +34,13 @@ from .numeric_conversion import int16_to_float
 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 .parallel import par_track
+from .parallel import track
 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
 from .ranges import expand_range
-from .read_mixture_data import read_mixture_data
 from .read_predict_data import read_predict_data
 from .reshape import get_input_shape
 from .reshape import get_num_classes_from_predict

sonusai/utils/asl_p56.py

@@ -88,7 +88,7 @@ def asl_p56(audio: AudioT) -> float:
                 # Interpolate to find the asl_ms_log
                 asl_ms_log = _bin_interp(A_db[j], A_db[j - 1], C_db[j], C_db[j - 1], M, 0.5)
                 # This is the mean square value NOT the RMS
-                asl_msq = 10. ** (asl_ms_log / 10)
+                asl_msq = 10.0 ** (asl_ms_log / 10)
                 break
 
     return asl_msq

sonusai/utils/asr.py

@@ -1,6 +1,5 @@
+from collections.abc import Callable
 from dataclasses import dataclass
-from typing import Callable
-from typing import Optional
 
 from sonusai.mixture import AudioT
 
@@ -8,25 +7,25 @@ from sonusai.mixture import AudioT
 @dataclass(frozen=True)
 class ASRResult:
     text: str
-    confidence: Optional[float] = None
-    lang: Optional[str] = None
-    lang_prob: Optional[float] = None
-    duration: Optional[float] = None
-    num_segments: Optional[int] = None
-    asr_cpu_time: Optional[float] = None
+    confidence: float | None = None
+    lang: str | None = None
+    lang_prob: float | None = None
+    duration: float | None = None
+    num_segments: int | None = None
+    asr_cpu_time: float | None = None
 
 
 def get_available_engines() -> list[str]:
     from importlib import import_module
     from pkgutil import iter_modules
 
-    module = import_module('sonusai.utils.asr_functions')
-    engines = [method for method in dir(module) if not method.startswith('_')]
+    module = import_module("sonusai.utils.asr_functions")
+    engines = [method for method in dir(module) if not method.startswith("_")]
     for _, name, _ in iter_modules():
-        if name.startswith('sonusai_asr_'):
-            module = import_module(f'{name}.asr_functions')
+        if name.startswith("sonusai_asr_"):
+            module = import_module(f"{name}.asr_functions")
             for method in dir(module):
-                if not method.startswith('_'):
+                if not method.startswith("_"):
                     engines.append(method)
 
     return engines
@@ -36,19 +35,19 @@ def _asr_fn(engine: str) -> Callable[..., ASRResult]:
     from importlib import import_module
     from pkgutil import iter_modules
 
-    module = import_module('sonusai.utils.asr_functions')
+    module = import_module("sonusai.utils.asr_functions")
     for method in dir(module):
         if method == engine:
             return getattr(module, method)
 
     for _, name, _ in iter_modules():
-        if name.startswith('sonusai_asr_'):
-            module = import_module(f'{name}.asr_functions')
+        if name.startswith("sonusai_asr_"):
+            module = import_module(f"{name}.asr_functions")
             for method in dir(module):
                 if method == engine:
                     return getattr(module, method)
 
-    raise ValueError(f'engine {engine} not supported')
+    raise ValueError(f"engine {engine} not supported")
 
 
 def calc_asr(audio: AudioT | str, engine: str, **config) -> ASRResult:
@@ -69,3 +68,24 @@ def calc_asr(audio: AudioT | str, engine: str, **config) -> ASRResult:
         audio = copy(read_audio(audio))
 
     return _asr_fn(engine)(audio, **config)
+
+
+def validate_asr(engine: str, **config) -> None:
+    from importlib import import_module
+    from pkgutil import iter_modules
+
+    module = import_module("sonusai.utils.asr_functions")
+    for method in dir(module):
+        if method == engine:
+            getattr(module, method + "_validate")(**config)
+            return
+
+    for _, name, _ in iter_modules():
+        if name.startswith("sonusai_asr_"):
+            module = import_module(f"{name}.asr_functions")
+            for method in dir(module):
+                if method == engine:
+                    getattr(module, method + "_validate")(**config)
+                    return
+
+    raise ValueError(f"engine {engine} not supported")

sonusai/utils/asr_functions/__init__.py

@@ -1 +1,3 @@
+# ruff: noqa: F401
+
 from .aaware_whisper import aaware_whisper

sonusai/utils/asr_functions/aaware_whisper.py

@@ -2,6 +2,10 @@ from sonusai.mixture import AudioT
 from sonusai.utils import ASRResult
 
 
+def aaware_whisper_validate(**_config) -> None:
+    pass
+
+
 def aaware_whisper(audio: AudioT, **_config) -> ASRResult:
     import tempfile
     from math import exp
@@ -10,32 +14,34 @@ def aaware_whisper(audio: AudioT, **_config) -> ASRResult:
 
     import requests
 
-    from sonusai import SonusAIError
     from sonusai.utils import ASRResult
     from sonusai.utils import float_to_int16
     from sonusai.utils import write_audio
 
-    url = getenv('AAWARE_WHISPER_URL')
+    url = getenv("AAWARE_WHISPER_URL")
     if url is None:
-        raise SonusAIError(f'AAWARE_WHISPER_URL environment variable does not exist')
-    url += '/asr?task=transcribe&language=en&encode=true&output=json'
+        raise EnvironmentError("AAWARE_WHISPER_URL environment variable does not exist")
+    url += "/asr?task=transcribe&language=en&encode=true&output=json"
 
     with tempfile.TemporaryDirectory() as tmp:
-        file = join(tmp, 'asr.wav')
+        file = join(tmp, "asr.wav")
         write_audio(name=file, audio=float_to_int16(audio))
 
-        files = {'audio_file': (file, open(file, 'rb'), 'audio/wav')}
+        files = {"audio_file": (file, open(file, "rb"), "audio/wav")}  # noqa: SIM115
 
         try:
-            response = requests.post(url, files=files)
-            if not response.status_code == 200:
+            response = requests.post(url, files=files)  # noqa: S113
+            if response.status_code != 200:
                 if response.status_code == 422:
-                    raise SonusAIError(f'Validation error: {response.json()}')
-                raise SonusAIError(f'Invalid response: {response.status_code}')
+                    raise RuntimeError(f"Validation error: {response.json()}")  # noqa: TRY301
+                raise RuntimeError(f"Invalid response: {response.status_code}")  # noqa: TRY301
             result = response.json()
-            return ASRResult(text=result['text'], confidence=exp(float(result['segments'][0]['avg_logprob'])))
+            return ASRResult(
+                text=result["text"],
+                confidence=exp(float(result["segments"][0]["avg_logprob"])),
+            )
         except Exception as e:
-            raise SonusAIError(f'Aaware Whisper exception: {e.args}')
+            raise RuntimeError(f"Aaware Whisper exception: {e.args}") from e
 
 
 """

sonusai/utils/audio_devices.py

@@ -1,29 +1,29 @@
 import pyaudio
 
 
-def get_input_device_index_by_name(p: pyaudio.PyAudio, name: str = None) -> int:
+def get_input_device_index_by_name(p: pyaudio.PyAudio, name: str | None = None) -> int:
     info = p.get_host_api_info_by_index(0)
-    device_count = info.get('deviceCount')
+    device_count = info.get("deviceCount")
     for i in range(0, device_count):
         device_info = p.get_device_info_by_host_api_device_index(0, i)
         if name is None:
             device_name = None
         else:
-            device_name = device_info.get('name')
-        if name == device_name and device_info.get('maxInputChannels') > 0:
+            device_name = device_info.get("name")
+        if name == device_name and device_info.get("maxInputChannels") > 0:
             return i
 
-    raise ValueError(f'Could not find {name}')
+    raise ValueError(f"Could not find {name}")
 
 
 def get_input_devices(p: pyaudio.PyAudio) -> list[str]:
     names = []
     info = p.get_host_api_info_by_index(0)
-    device_count = info.get('deviceCount')
+    device_count = info.get("deviceCount")
     for i in range(0, device_count):
         device_info = p.get_device_info_by_host_api_device_index(0, i)
-        device_name = device_info.get('name')
-        if device_info.get('maxInputChannels') > 0:
+        device_name = device_info.get("name")
+        if device_info.get("maxInputChannels") > 0:
             names.append(device_name)
 
     return names
@@ -31,11 +31,11 @@ def get_input_devices(p: pyaudio.PyAudio) -> list[str]:
 
 def get_default_input_device(p: pyaudio.PyAudio) -> str:
     info = p.get_host_api_info_by_index(0)
-    device_count = info.get('deviceCount')
+    device_count = info.get("deviceCount")
     for i in range(0, device_count):
         device_info = p.get_device_info_by_host_api_device_index(0, i)
-        device_name = device_info.get('name')
-        if device_info.get('maxInputChannels') > 0:
+        device_name = device_info.get("name")
+        if device_info.get("maxInputChannels") > 0:
             return device_name
 
-    raise ValueError('No input audio devices found')
+    raise ValueError("No input audio devices found")

sonusai/utils/braced_glob.py

@@ -1,10 +1,8 @@
-from typing import Generator
+from collections.abc import Generator
 from typing import LiteralString
-from typing import Optional
-from typing import Set
 
 
-def expand_braces(text: LiteralString | str | bytes, seen: Optional[Set[str]] = None) -> Generator[str, None, None]:
+def expand_braces(text: LiteralString | str | bytes, seen: set[str] | None = None) -> Generator[str, None, None]:
     """Brace-expansion pre-processing for glob.
 
     Expand all the braces, then run glob on each of the results.
@@ -20,8 +18,8 @@ def expand_braces(text: LiteralString | str | bytes, seen: Optional[Set[str]] =
     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]
+    spans = [m.span() for m in re.finditer(r"\{[^{}]*}", text)][::-1]
+    alts = [text[start + 1 : stop - 1].split(",") for start, stop in spans]
 
     if len(spans) == 0:
         if text not in seen:
@@ -30,9 +28,9 @@ def expand_braces(text: LiteralString | str | bytes, seen: Optional[Set[str]] =
     else:
         for combo in itertools.product(*alts):
             replaced = list(text)
-            for (start, stop), replacement in zip(spans, combo):
+            for (start, stop), replacement in zip(spans, combo, strict=False):
                 replaced[start:stop] = replacement
-            yield from expand_braces(''.join(replaced), seen)
+            yield from expand_braces("".join(replaced), seen)
 
 
 def braced_glob(pathname: LiteralString | str | bytes, recursive: bool = False) -> list[str]:

sonusai/utils/calculate_input_shape.py

@@ -1,7 +1,4 @@
-def calculate_input_shape(feature: str,
-                          flatten: bool = False,
-                          timesteps: int = 0,
-                          add1ch: bool = False) -> list[int]:
+def calculate_input_shape(feature: str, flatten: bool = False, timesteps: int = 0, add1ch: bool = False) -> list[int]:
     """
     Calculate input shape given feature and user-specified reshape parameters.
 

sonusai/utils/compress.py

@@ -6,7 +6,7 @@ def power_compress(feature: AudioF) -> AudioF:
 
     mag = np.abs(feature)
     phase = np.angle(feature)
-    mag = mag ** 0.3
+    mag = mag**0.3
     real_compress = mag * np.cos(phase)
     imag_compress = mag * np.sin(phase)
 
@@ -18,7 +18,7 @@ def power_uncompress(feature: AudioF) -> AudioF:
 
     mag = np.abs(feature)
     phase = np.angle(feature)
-    mag = mag ** (1. / 0.3)
+    mag = mag ** (1.0 / 0.3)
     real_uncompress = mag * np.cos(phase)
     imag_uncompress = mag * np.sin(phase)
 

sonusai/utils/convert_string_to_number.py

@@ -1,8 +1,6 @@
 def convert_string_to_number(string: str) -> float | int | str:
     try:
         result = float(string)
-        if result == int(result):
-            return int(result)
-        return result
+        return int(result) if result == int(result) else result
     except ValueError:
         return string

sonusai/utils/create_timestamp.py

@@ -2,4 +2,4 @@ def create_timestamp() -> str:
     """Create a timestamp."""
     from datetime import datetime
 
-    return datetime.now().strftime('%Y%m%d-%H%M%S')
+    return datetime.now().strftime("%Y%m%d-%H%M%S")

sonusai/utils/create_ts_name.py

@@ -6,9 +6,9 @@ def create_ts_name(base: str) -> str:
     ts = datetime.now()
 
     # First try just date
-    dir_name = base + '-' + ts.strftime('%Y%m%d')
+    dir_name = base + "-" + ts.strftime("%Y%m%d")
     if exists(dir_name):
         # add hour-min-sec if necessary
-        dir_name = base + '-' + ts.strftime('%Y%m%d-%H%M%S')
+        dir_name = base + "-" + ts.strftime("%Y%m%d-%H%M%S")
 
     return dir_name

sonusai/utils/dataclass_from_dict.py

@@ -4,6 +4,6 @@ def dataclass_from_dict(klass, dikt):
         field_types = klass.__annotations__
         return klass(**{f: dataclass_from_dict(field_types[f], dikt[f]) for f in dikt})
     except AttributeError:
-        if isinstance(dikt, (tuple, list)):
+        if isinstance(dikt, tuple | list):
             return [dataclass_from_dict(klass.__args__[0], f) for f in dikt]
         return dikt

sonusai/utils/docstring.py

@@ -3,7 +3,7 @@ def trim_docstring(docstring: str) -> str:
     from sys import maxsize
 
     if not docstring:
-        return ''
+        return ""
 
     # Convert tabs to spaces (following the normal Python rules)
     # and split into a list of lines:
@@ -27,7 +27,7 @@ def trim_docstring(docstring: str) -> str:
         trimmed.pop(0)
 
     # Return a single string
-    return '\n'.join(trimmed)
+    return "\n".join(trimmed)
 
 
 def add_commands_to_docstring(docstring: str, plugin_docstrings: list[str]) -> str:
@@ -36,8 +36,8 @@ def add_commands_to_docstring(docstring: str, plugin_docstrings: list[str]) -> s
 
     lines = docstring.splitlines()
 
-    start = lines.index('The sonusai commands are:')
-    end = lines.index('', start)
+    start = lines.index("The sonusai commands are:")
+    end = lines.index("", start)
 
     commands = sonusai.commands_doc.splitlines()
     for plugin_docstring in plugin_docstrings:
@@ -45,6 +45,6 @@ def add_commands_to_docstring(docstring: str, plugin_docstrings: list[str]) -> s
     commands.sort()
     commands = list(filter(None, commands))
 
-    lines = lines[:start + 1] + commands + lines[end:]
+    lines = lines[: start + 1] + commands + lines[end:]
 
-    return '\n'.join(lines)
+    return "\n".join(lines)

sonusai/utils/energy_f.py

@@ -1,12 +1,15 @@
-from sonusai import ForwardTransform
+from pyaaware import ForwardTransform
+
 from sonusai.mixture import AudioF
 from sonusai.mixture import AudioT
 from sonusai.mixture import EnergyF
 
 
-def compute_energy_f(frequency_domain: AudioF = None,
-                     time_domain: AudioT = None,
-                     transform: ForwardTransform = None) -> EnergyF:
+def compute_energy_f(
+    frequency_domain: AudioF | None = None,
+    time_domain: AudioT | None = None,
+    transform: ForwardTransform | None = None,
+) -> EnergyF:
     """Compute the energy in each bin
 
     Must provide either frequency domain or time domain input. If time domain input is provided, must also provide
@@ -19,13 +22,12 @@ def compute_energy_f(frequency_domain: AudioF = None,
     """
     import numpy as np
     import torch
-    from sonusai import SonusAIError
 
     if frequency_domain is None:
         if time_domain is None:
-            raise SonusAIError('Must provide time or frequency domain input')
+            raise ValueError("Must provide time or frequency domain input")
         if transform is None:
-            raise SonusAIError('Must provide ForwardTransform object')
+            raise ValueError("Must provide ForwardTransform object")
 
         frequency_domain = transform.execute_all(torch.from_numpy(time_domain))[0].numpy()