diff --git a/espnet2/bin/asr_inference.py b/espnet2/bin/asr_inference.py
index 810f8714a9f..b6f86e6aa43 100755
--- a/espnet2/bin/asr_inference.py
+++ b/espnet2/bin/asr_inference.py
@@ -123,11 +123,9 @@ def __init__(
 
         if quantize_asr_model:
             logging.info("Use quantized asr model for decoding.")
-            
+
             asr_model = torch.quantization.quantize_dynamic(
-                asr_model, 
-                qconfig_spec=quantize_modules, 
-                dtype=quantize_dtype
+                asr_model, qconfig_spec=quantize_modules, dtype=quantize_dtype
             )
 
         decoder = asr_model.decoder
@@ -150,9 +148,7 @@ def __init__(
                 logging.info("Use quantized lm for decoding.")
 
                 lm = torch.quantization.quantize_dynamic(
-                    lm,
-                    qconfig_spec=quantize_modules,
-                    dtype=quantize_dtype
+                    lm, qconfig_spec=quantize_modules, dtype=quantize_dtype
                 )
 
             scorers["lm"] = lm.lm
diff --git a/test/espnet2/bin/test_asr_inference.py b/test/espnet2/bin/test_asr_inference.py
index 1946e6ce475..c64a0c774eb 100644
--- a/test/espnet2/bin/test_asr_inference.py
+++ b/test/espnet2/bin/test_asr_inference.py
@@ -87,10 +87,10 @@ def test_Speech2Text(asr_config_file, lm_config_file):
 @pytest.mark.execution_timeout(5)
 def test_Speech2Text_quantized(asr_config_file, lm_config_file):
     speech2text = Speech2Text(
-        asr_train_config=asr_config_file, 
-        lm_train_config=lm_config_file, 
+        asr_train_config=asr_config_file,
+        lm_train_config=lm_config_file,
         beam_size=1,
-        quantize_asr_model=True, 
+        quantize_asr_model=True,
         quantize_lm=True,
     )
     speech = np.random.randn(100000)