Merge pull request #2 from joelmathewthomas/feature/input-and-preprocessing

Input & Preprocessing Stage: Implement Audio File Reading and Format Validation

.gitignore

@@ -1,2 +1,16 @@
+# Ignore Python bytecode cache files
+__pycache__/
+
+# Ignore pytest cache
+.pytest_cache/
+
+# Ignore virtual environment folder
 venv/
-samples/
+
+# Ignore other common files
+*.pyc
+*.pyo
+*.pyd
+
+# Ignore VSCode config
+.vscode/

pytest.ini

@@ -0,0 +1,2 @@
+[pytest]
+pythonpath = . src

requirements.txt

@@ -0,0 +1,28 @@
+audioread==3.0.1
+certifi==2024.12.14
+cffi==1.17.1
+charset-normalizer==3.4.0
+decorator==5.1.1
+idna==3.10
+iniconfig==2.0.0
+joblib==1.4.2
+lazy_loader==0.4
+librosa==0.10.2.post1
+llvmlite==0.43.0
+msgpack==1.1.0
+numba==0.60.0
+numpy==2.0.2
+packaging==24.2
+platformdirs==4.3.6
+pluggy==1.5.0
+pooch==1.8.2
+pycparser==2.22
+pytest==8.3.4
+requests==2.32.3
+scikit-learn==1.6.0
+scipy==1.14.1
+soundfile==0.12.1
+soxr==0.5.0.post1
+threadpoolctl==3.5.0
+typing_extensions==4.12.2
+urllib3==2.3.0

src/input/__init__.py

@@ -0,0 +1,12 @@
+# __init__.py
+
+import logging
+from datetime import datetime
+
+# Configure logging
+logging.basicConfig(
+    format='%(asctime)s : %(message)s',
+    level = logging.INFO
+)
+
+logging.info("freq-split-enhance/input package has been imported.")

src/input/file_reader.py

@@ -0,0 +1,21 @@
+import os
+import librosa
+
+def read_audio(file_path):
+    """
+    Reads an audio file and returns the audio time series and sampling rate.
+
+    Args:
+        file_path (str): Path to the audio file.
+
+    Returns:
+        tuple: audio_time_series (numpy.ndarray), sampling_rate (int)
+    """
+
+    if not os.path.exists(file_path):
+        raise FileNotFoundError(f"File not found: {file_path}")
+    try:
+        audio, sr = librosa.load(file_path, sr=None) # Load with original sampling rate.
+        return audio, sr
+    except Exception as e:
+        raise RuntimeError(f"Error reading the audio file: {e}")

src/input/format_checker.py

@@ -0,0 +1,15 @@
+import mimetypes
+
+def is_supported_format(file_path):
+    """
+    Checks if the audio file is in a supported format.
+
+    Args:
+        file_path (str): Path to the audio file.
+
+    Returns:
+        bool: True if supported, False otherwise.
+    """
+    supported_formats = ["audio/mpeg", "audio/wav", "audio/x-aiff", "audio/x-wav", ...]
+    mime_type, _ = mimetypes.guess_type(file_path)
+    return mime_type in supported_formats

src/preprocessing/__init__.py

@@ -0,0 +1,12 @@
+# __init__.py
+
+import logging
+from datetime import datetime
+
+# Configure logging
+logging.basicConfig(
+    format='%(asctime)s : %(message)s',
+    level = logging.INFO
+)
+
+logging.info("freq-split-enhance/preprocessing package has been imported.")

src/preprocessing/normalize.py

@@ -0,0 +1,14 @@
+import librosa
+import numpy as np
+
+def normalize_audio(audio: np.ndarray) -> np.ndarray:
+    """
+    Normalize the audio to a range of [-1, 1].
+
+    Args:
+    - audio (np.ndarray): The audio time series to normalize.
+
+    Returns:
+    - np.ndarray: The normalized audio time series.
+    """
+    return librosa.util.normalize(audio)

src/preprocessing/trim.py

@@ -0,0 +1,17 @@
+import librosa
+import numpy as np
+
+def trim_audio(audio:np.ndarray, sr:int) -> np.ndarray:
+    """
+    Trim leading and trailing silence from the audio.
+
+    Args:
+    - audio (np.ndarray): The audio time series.
+    - sr (int): The sample rate of the audio.
+
+    Returns:
+    - np.ndarray: The trimmed audio time series.
+    """
+
+    audio_trimmed, _ = librosa.effects.trim(audio)
+    return audio_trimmed

tests/test_input.py

@@ -0,0 +1,13 @@
+import pytest
+from  src.input.file_reader import read_audio
+from src.input.format_checker import is_supported_format
+
+def test_read_audio():
+    file_path = "samples/cafe_crowd_talk.aiff"
+    audio, sr = read_audio(file_path)
+    assert len(audio) > 0
+    assert sr > 0
+
+def test_is_supported_format():
+    assert is_supported_format("samples/cafe_crowd_talk.aiff") == True
+    assert is_supported_format("samples/unsupported_file.txt") == False

tests/test_preprocessing.py

@@ -0,0 +1,20 @@
+import pytest
+import librosa
+from src.preprocessing.normalize import normalize_audio
+from src.preprocessing.trim import trim_audio
+from src.input.file_reader import read_audio
+
+def test_normalize_audio():
+    file_path = "samples/cafe_crowd_talk.aiff"
+    audio, _ = read_audio(file_path)
+    normalized_audio = normalize_audio(audio)
+
+    assert normalized_audio.max() <= 1.0
+    assert normalized_audio.min() >= -1.0
+
+def test_trim_audio():
+    file_path = "samples/cafe_crowd_talk.aiff"
+    audio, sr = read_audio(file_path)
+    trimmed_audio = trim_audio(audio, sr)
+
+    assert len(trimmed_audio) <= len(audio)