Update tasks/audio.py

tasks/audio.py

@@ -1,13 +1,12 @@
-import librosa
-import joblib
-import numpy as np
-
 from fastapi import APIRouter
 from datetime import datetime
 from datasets import load_dataset
 from sklearn.metrics import accuracy_score
 import random
 import os
+import librosa
+import joblib
+import numpy as np
 
 from .utils.evaluation import AudioEvaluationRequest
 from .utils.emissions import tracker, clean_emissions_data, get_space_info
@@ -17,7 +16,7 @@ load_dotenv()
 
 router = APIRouter()
 
-DESCRIPTION = "Random Baseline"
+DESCRIPTION = "LGBM Classifier Baseline on Mel-frequency cepstral coefficients"
 ROUTE = "/audio"
 
 
@@ -28,8 +27,7 @@ async def evaluate_audio(request: AudioEvaluationRequest):
     """
     Evaluate audio classification for rainforest sound detection.
     
-    Current Model: Random Baseline
-    - Makes random predictions from the label space (0-1)
+    Current Model: LGBM
     - Used as a baseline for comparison
     """
     # Get space info
@@ -57,15 +55,15 @@ async def evaluate_audio(request: AudioEvaluationRequest):
     # Update the code below to replace the random baseline by your model inference within the inference pass where the energy consumption and emissions are tracked.
     #--------------------------------------------------------------------------------------------   
     
-    def extract_features(example, sampling_rate):
-        audio_array = example['audio']['array']
-        mfcc = librosa.feature.mfcc(y=audio_array, sr=sampling_rate, n_mfcc=5)
+    def compute_mfcc(row, sr):
+        audio_array = row['audio']['array']
+        mfcc = librosa.feature.mfcc(y=audio_array, sr=sr, n_mfcc=5)
         return np.mean(mfcc, axis=1)
 
-    def predict_new_audio(model, dataset, sampling_rate):
-        features_list = [extract_features(example, sampling_rate) for example in dataset]
-        features_array = np.vstack(features_list)
-        predictions = model.predict(features_array)
+    def predict_new_audio(model, dataset, sr):
+        list_mfcc = [extract_features(row, sr) for row in dataset]
+        array_mfcc = np.vstack(list_mfcc)
+        predictions = model.predict(array_mfcc)
         return predictions
 
     model_filename = "models/lightgbm_baseline_87_acc.pkl"