From d43f77408b8f3e169c59dfb6b6d82e45e6b91714 Mon Sep 17 00:00:00 2001
From: 游雁 <zhifu.gzf@alibaba-inc.com>
Date: 星期二, 11 六月 2024 19:19:06 +0800
Subject: [PATCH] decoding

---
 funasr/auto/auto_model.py |   16 ++++++++++++----
 1 files changed, 12 insertions(+), 4 deletions(-)

diff --git a/funasr/auto/auto_model.py b/funasr/auto/auto_model.py
index 97eb325..bbaf657 100644
--- a/funasr/auto/auto_model.py
+++ b/funasr/auto/auto_model.py
@@ -42,8 +42,9 @@
     filelist = [".scp", ".txt", ".json", ".jsonl", ".text"]
 
     chars = string.ascii_letters + string.digits
-    if isinstance(data_in, str) and data_in.startswith("http"):  # url
-        data_in = download_from_url(data_in)
+    if isinstance(data_in, str):
+        if data_in.startswith("http://") or data_in.startswith("https://"):  # url
+            data_in = download_from_url(data_in)
 
     if isinstance(data_in, str) and os.path.exists(
         data_in
@@ -211,7 +212,6 @@
         deep_update(model_conf, kwargs.get("model_conf", {}))
         deep_update(model_conf, kwargs)
         model = model_class(**model_conf, vocab_size=vocab_size)
-        model.to(device)
 
         # init_param
         init_param = kwargs.get("init_param", None)
@@ -232,6 +232,9 @@
         # fp16
         if kwargs.get("fp16", False):
             model.to(torch.float16)
+        elif kwargs.get("bf16", False):
+            model.to(torch.bfloat16)
+        model.to(device)
         return model, kwargs
 
     def __call__(self, *args, **cfg):
@@ -284,7 +287,7 @@
             with torch.no_grad():
                 res = model.inference(**batch, **kwargs)
                 if isinstance(res, (list, tuple)):
-                    results = res[0]
+                    results = res[0] if len(res) > 0 else [{"text": ""}]
                     meta_data = res[1] if len(res) > 1 else {}
             time2 = time.perf_counter()
 
@@ -358,6 +361,7 @@
             results_sorted = []
 
             if not len(sorted_data):
+                results_ret_list.append({"key": key, "text": "", "timestamp": []})
                 logging.info("decoding, utt: {}, empty speech".format(key))
                 continue
 
@@ -425,6 +429,10 @@
             #                      f"time_speech_total_per_sample: {time_speech_total_per_sample: 0.3f}, "
             #                      f"time_escape_total_per_sample: {time_escape_total_per_sample:0.3f}")
 
+            if len(results_sorted) != n:
+                results_ret_list.append({"key": key, "text": "", "timestamp": []})
+                logging.info("decoding, utt: {}, empty result".format(key))
+                continue
             restored_data = [0] * n
             for j in range(n):
                 index = sorted_data[j][1]

--
Gitblit v1.9.1