fix: Don't load DeepSpeed if use_deepspeed is False

- A recent change made DeepSpeed optional (off by default), but the code was still trying to load DeepSpeed even when `use_deepspeed = False`. This means users would still have a big startup slowdown and a lot of error messages if their DeepSpeed module isn't working (usually because it's not able to compile itself on their machines).

- We now only load DeepSpeed if the user requested it.

- Translated the DeepSpeed error message to English, since all other errors in the same function were already English.

indextts/infer_v2.py

@@ -47,7 +47,7 @@ class IndexTTS2:
             use_fp16 (bool): whether to use fp16.
             device (str): device to use (e.g., 'cuda:0', 'cpu'). If None, it will be set automatically based on the availability of CUDA or MPS.
             use_cuda_kernel (None | bool): whether to use BigVGan custom fused activation CUDA kernel, only for CUDA device.
-            use_deepspeed (bool): whether to use deepspeed or not.
+            use_deepspeed (bool): whether to use DeepSpeed or not.
         """
         if device is not None:
             self.device = device
@@ -88,12 +88,12 @@ class IndexTTS2:
             self.gpt.eval()
         print(">> GPT weights restored from:", self.gpt_path)
 
-        try:
-            import deepspeed
-        except (ImportError, OSError, CalledProcessError) as e:
-            if use_deepspeed:
-                print(f">> DeepSpeed加载失败，回退到标准推理: {e}")
-            use_deepspeed = False
+        if use_deepspeed:
+            try:
+                import deepspeed
+            except (ImportError, OSError, CalledProcessError) as e:
+                use_deepspeed = False
+                print(f">> Failed to load DeepSpeed. Falling back to normal inference. Error: {e}")
 
         self.gpt.post_init_gpt2_config(use_deepspeed=use_deepspeed, kv_cache=True, half=self.use_fp16)
 

webui.py

@@ -25,7 +25,7 @@ parser.add_argument("--port", type=int, default=7860, help="Port to run the web
 parser.add_argument("--host", type=str, default="0.0.0.0", help="Host to run the web UI on")
 parser.add_argument("--model_dir", type=str, default="./checkpoints", help="Model checkpoints directory")
 parser.add_argument("--fp16", action="store_true", default=False, help="Use FP16 for inference if available")
-parser.add_argument("--use_deepspeed", action="store_true", default=False, help="Use Deepspeed to accelerate if available")
+parser.add_argument("--use_deepspeed", action="store_true", default=False, help="Use DeepSpeed to accelerate if available")
 parser.add_argument("--cuda_kernel", action="store_true", default=False, help="Use CUDA kernel for inference if available")
 parser.add_argument("--gui_seg_tokens", type=int, default=120, help="GUI: Max tokens per generation segment")
 cmd_args = parser.parse_args()