diff --git a/cosyvoice/llm/llm.py b/cosyvoice/llm/llm.py
index 5d6e4db..e5a961c 100644
--- a/cosyvoice/llm/llm.py
+++ b/cosyvoice/llm/llm.py
@@ -345,21 +345,22 @@ class Qwen2LM(TransformerLM):
             vllm_codec_engine.add_request(request_id,
                                         {"prompt_embeds": lm_input.squeeze(0).to(torch.bfloat16).to(device)},
                                         sampling_params)
-            ## generator
-            out_token_ids = []
             while True:
+                speech_token_break = False
                 request_outputs: List[RequestOutput] = vllm_codec_engine.step()
                 for request_output in request_outputs:
                     if str(request_output.request_id) != str(request_id):
                         continue
-                    if not request_output.finished:
-                        # print(f"Partial request output: {request_output}")
-                        out_token = list(request_output.outputs[0].token_ids)[-1]
-                        yield out_token
-                        out_token_ids.append(out_token)
-                    else:
+                    # print(f"request output: {request_output}")
+                    top_ids = list(request_output.outputs[0].token_ids)[-1]
+                    if top_ids == self.speech_token_size:
+                        speech_token_break = True
                         break
-                if not vllm_codec_engine.has_unfinished_requests():
+                    if top_ids > self.speech_token_size:
+                        continue
+                    yield top_ids
+
+                if not vllm_codec_engine.has_unfinished_requests() or speech_token_break:
                     break
 
     @torch.inference_mode()