Merge pull request #2129 from lym0302/onnx_gpu

[server]specify id

paddlespeech/server/utils/onnx_infer.py

@@ -30,7 +30,9 @@ def get_sess(model_path: Optional[os.PathLike]=None, sess_conf: dict=None):
     # "gpu:0"
     providers = ['CPUExecutionProvider']
     if "gpu" in sess_conf.get("device", ""):
-        providers = ['CUDAExecutionProvider']
+        device_id = int(sess_conf["device"].split(":")[1])
+        providers = [('CUDAExecutionProvider', {'device_id': device_id})]
+
         # fastspeech2/mb_melgan can't use trt now!
         if sess_conf.get("use_trt", 0):
             providers = ['TensorrtExecutionProvider']