Merge pull request #1065 from LittleChenCc/develop

[ST] activate eval mode in translation demo function

examples/ted_en_zh/st0/conf/transformer_mtl_noam.yaml

@@ -86,7 +86,7 @@ training:
   scheduler_conf:
     warmup_steps: 25000
     lr_decay: 1.0
-  log_interval: 5
+  log_interval: 50
   checkpoint:
     kbest_n: 50
     latest_n: 5

paddlespeech/s2t/exps/u2_st/model.py

@@ -432,6 +432,7 @@ class U2STTester(U2STTrainer):
         """"E2E translation from extracted audio feature"""
         cfg = self.config.decoding
         text_feature = self.test_loader.collate_fn.text_feature
+        self.model.eval()
 
         hyps = self.model.decode(
             audio,