diff --git a/src/preprocessor/soundpreprocessor.py b/src/preprocessor/soundpreprocessor.py
index a3ce688..adcc9c4 100644
--- a/src/preprocessor/soundpreprocessor.py
+++ b/src/preprocessor/soundpreprocessor.py
@@ -32,7 +32,13 @@ class SoundPreProcessor(AbcPreProcessor):
         self._temp_means_name = self._temp_model_name + "MEANS"
 
         logging.debug("Fetching model info...")
-        r = requests.get(f"{config.API_URL}/model/svm/$default/details")
+
+        if config.SVM_MODEL_ID:
+            model_id_to_get = config.SVM_MODEL_ID
+        else:
+            model_id_to_get = '$default'
+
+        r = requests.get(f"{config.API_URL}/model/svm/{model_id_to_get}/details")
         r.raise_for_status()
 
         self._model_details = r.json()
@@ -97,7 +103,7 @@ class SoundPreProcessor(AbcPreProcessor):
 
         logging.info("Running classification...")
 
-        target_id = self._classes.index('chirp')  # Might raise ValueError
+        target_id = self._classes.index(config.SVM_TARGET_CLASS_NAME)  # Might raise ValueError
 
         feature_vector = (mid_features - self._mean) / self._std
         class_id, probability = classifier_wrapper(self._classifier, self._model_details['type'], feature_vector)
diff --git a/src/utils/config.py b/src/utils/config.py
index 0e9f44e..67d5328 100644
--- a/src/utils/config.py
+++ b/src/utils/config.py
@@ -27,7 +27,7 @@ MQTT_PORT = int(os.getenv("GUARD_MQTT_PORT", 1883))
 MQTT_USERNAME = os.getenv("GUARD_MQTT_USERNAME", None)
 MQTT_PASSWORD = os.getenv("GUARD_MQTT_PASSWORD", None)
 
-TARGET_NAME = os.environ.get("TARGET_CLASS_NAME")
-MODEL_ID = os.environ.get("MODEL_ID")
+SVM_TARGET_CLASS_NAME = os.environ.get("SVM_TARGET_CLASS_NAME", "chirp")
+SVM_MODEL_ID = os.environ.get("SVM_MODEL_ID")
 
 API_URL = os.environ.get("API_URL", "http://localhost:8080")
\ No newline at end of file