diff --git a/indextts/infer_v2.py b/indextts/infer_v2.py
index a12105b..cf66d36 100644
--- a/indextts/infer_v2.py
+++ b/indextts/infer_v2.py
@@ -413,6 +413,13 @@ class IndexTTS2:
 
         # 如果参考音频改变了，才需要重新生成, 提升速度
         if self.cache_spk_cond is None or self.cache_spk_audio_prompt != spk_audio_prompt:
+            if self.cache_spk_cond is not None:
+                self.cache_spk_cond = None
+                self.cache_s2mel_style = None
+                self.cache_s2mel_prompt = None
+                self.cache_mel = None
+                if torch.cuda.is_available():
+                    torch.cuda.empty_cache()
             audio, sr = librosa.load(spk_audio_prompt)
             audio = torch.tensor(audio).unsqueeze(0)
             audio_22k = torchaudio.transforms.Resample(sr, 22050)(audio)
@@ -465,6 +472,10 @@ class IndexTTS2:
             emovec_mat = emovec_mat.unsqueeze(0)
 
         if self.cache_emo_cond is None or self.cache_emo_audio_prompt != emo_audio_prompt:
+            if self.cache_emo_cond is not None:
+                self.cache_emo_cond = None
+                if torch.cuda.is_available():
+                    torch.cuda.empty_cache()
             emo_audio, _ = librosa.load(emo_audio_prompt, sr=16000)
             emo_inputs = self.extract_features(emo_audio, sampling_rate=16000, return_tensors="pt")
             emo_input_features = emo_inputs["input_features"]