From adcee8828ef5d78b575043954deb662a35e318f7 Mon Sep 17 00:00:00 2001
From: huangmingming <huangmingming@deepscience.cn>
Date: 星期一, 30 一月 2023 16:02:54 +0800
Subject: [PATCH] update the minimum size of audio

---
 funasr/bin/asr_inference_paraformer_vad_punc.py |   23 +++++++++++++++++------
 1 files changed, 17 insertions(+), 6 deletions(-)

diff --git a/funasr/bin/asr_inference_paraformer_vad_punc.py b/funasr/bin/asr_inference_paraformer_vad_punc.py
index 4b5b316..7a539e4 100644
--- a/funasr/bin/asr_inference_paraformer_vad_punc.py
+++ b/funasr/bin/asr_inference_paraformer_vad_punc.py
@@ -3,6 +3,7 @@
 import logging
 import sys
 import time
+import json
 from pathlib import Path
 from typing import Optional
 from typing import Sequence
@@ -100,10 +101,13 @@
         # logging.info("asr_train_args: {}".format(asr_train_args))
         asr_model.to(dtype=getattr(torch, dtype)).eval()
 
-        ctc = CTCPrefixScorer(ctc=asr_model.ctc, eos=asr_model.eos)
+        if asr_model.ctc != None:
+            ctc = CTCPrefixScorer(ctc=asr_model.ctc, eos=asr_model.eos)
+            scorers.update(
+                ctc=ctc
+            )
         token_list = asr_model.token_list
         scorers.update(
-            ctc=ctc,
             length_bonus=LengthBonus(len(token_list)),
         )
 
@@ -171,7 +175,7 @@
         self.converter = converter
         self.tokenizer = tokenizer
         is_use_lm = lm_weight != 0.0 and lm_file is not None
-        if ctc_weight == 0.0 and not is_use_lm:
+        if (ctc_weight == 0.0 or asr_model.ctc == None) and not is_use_lm:
             beam_search = None
         self.beam_search = beam_search
         logging.info(f"Beam_search: {self.beam_search}")
@@ -632,9 +636,16 @@
                         text_postprocessed, time_stamp_postprocessed, word_lists = postprocessed_result[0], \
                                                                                    postprocessed_result[1], \
                                                                                    postprocessed_result[2]
-                        text_postprocessed_punc, punc_id_list = text2punc(word_lists, 20)
-                        text_postprocessed_punc_time_stamp = "predictions: {}  time_stamp: {}".format(
-                            text_postprocessed_punc, time_stamp_postprocessed)
+                        if len(word_lists) > 0: 
+                            text_postprocessed_punc, punc_id_list = text2punc(word_lists, 20)
+                            text_postprocessed_punc_time_stamp = json.dumps({"predictions": text_postprocessed_punc,
+                                                                             "time_stamp": time_stamp_postprocessed},
+                                                                            ensure_ascii=False)
+                        else:
+                            text_postprocessed_punc = ""
+                            punc_id_list = []
+                            text_postprocessed_punc_time_stamp = ""
+                            
                     else:
                         text_postprocessed = ""
                         time_stamp_postprocessed = ""

--
Gitblit v1.9.1