fix a bug

Author: Jackwaterveg

paddlespeech/s2t/exps/deepspeech2/model.py

@@ -413,19 +413,19 @@ def test(self):
 
     def compute_result_transcripts(self, audio, audio_len):
         if self.args.model_type == "online":
-            output_probs, output_lens, trans = self.static_forward_online(
+            output_probs, output_lens, trans_batch = self.static_forward_online(
                 audio, audio_len, decoder_chunk_size=1)
-            result_transcripts = trans[-1:]
+            result_transcripts = [trans[-1] for trans in trans_batch]
         elif self.args.model_type == "offline":
-            batch_size = output_probs.shape[0]
-            self.model.decoder.reset_decoder(batch_size = batch_size)
             output_probs, output_lens = self.static_forward_offline(audio,
                                                                     audio_len)
+            batch_size = output_probs.shape[0]
+            self.model.decoder.reset_decoder(batch_size=batch_size)
 
             self.model.decoder.next(output_probs, output_lens)
 
             trans_best, trans_beam = self.model.decoder.decode()
-            
+
             result_transcripts = trans_best
 
         else:
@@ -485,7 +485,7 @@ def static_forward_online(self, audio, audio_len,
         x_list = np.split(x_batch, batch_size, axis=0)
         x_len_list = np.split(x_len_batch, batch_size, axis=0)
 
-        trans = []
+        trans_batch = []
         for x, x_len in zip(x_list, x_len_list):
             if self.args.enable_auto_log is True:
                 self.autolog.times.start()
@@ -518,14 +518,14 @@ def static_forward_online(self, audio, audio_len,
             h_box_handle = self.predictor.get_input_handle(input_names[2])
             c_box_handle = self.predictor.get_input_handle(input_names[3])
 
-            trans_chunk_list = []
+            trans = []
             probs_chunk_list = []
             probs_chunk_lens_list = []
             if self.args.enable_auto_log is True:
                 # record the model preprocessing time
                 self.autolog.times.stamp()
-            
-            self.model.decoder.reset_decoder(batch_size = 1)
+
+            self.model.decoder.reset_decoder(batch_size=1)
             for i in range(0, num_chunk):
                 start = i * chunk_stride
                 end = start + chunk_size
@@ -569,8 +569,7 @@ def static_forward_online(self, audio, audio_len,
                 probs_chunk_lens_list.append(output_chunk_lens)
                 trans_best, trans_beam = self.model.decoder.decode()
                 trans.append(trans_best[0])
-            
-
+            trans_batch.append(trans)
             output_probs = np.concatenate(probs_chunk_list, axis=1)
             output_lens = np.sum(probs_chunk_lens_list, axis=0)
             vocab_size = output_probs.shape[2]
@@ -592,7 +591,7 @@ def static_forward_online(self, audio, audio_len,
                 self.autolog.times.end()
         output_probs = np.concatenate(output_probs_list, axis=0)
         output_lens = np.concatenate(output_lens_list, axis=0)
-        return output_probs, output_lens, trans
+        return output_probs, output_lens, trans_batch
 
     def static_forward_offline(self, audio, audio_len):
         """