Update app.py
Browse files
app.py
CHANGED
@@ -53,7 +53,13 @@ class ImageCaptionTool(Tool):
|
|
53 |
|
54 |
def forward(self, image):
|
55 |
prompt = "Please describe the content of this picture in detail."
|
56 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
57 |
|
58 |
class AudioToTextTool(Tool):
|
59 |
name = "audio-to-text"
|
@@ -72,7 +78,12 @@ class AudioToTextTool(Tool):
|
|
72 |
|
73 |
def forward(self, audio):
|
74 |
prompt = "Please transcribe this audio content into text."
|
75 |
-
|
|
|
|
|
|
|
|
|
|
|
76 |
|
77 |
# --- Basic Agent Definition ---
|
78 |
# ----- THIS IS WERE YOU CAN BUILD WHAT YOU WANT ------
|
|
|
53 |
|
54 |
def forward(self, image):
|
55 |
prompt = "Please describe the content of this picture in detail."
|
56 |
+
result = self.model(prompt, images=[image])
|
57 |
+
# 兼容AgentText等包装类型,确保返回str
|
58 |
+
if hasattr(result, "to_raw"):
|
59 |
+
return result.to_raw()
|
60 |
+
if hasattr(result, "value"):
|
61 |
+
return result.value
|
62 |
+
return str(result)
|
63 |
|
64 |
class AudioToTextTool(Tool):
|
65 |
name = "audio-to-text"
|
|
|
78 |
|
79 |
def forward(self, audio):
|
80 |
prompt = "Please transcribe this audio content into text."
|
81 |
+
result = self.model(prompt, audios=[audio])
|
82 |
+
if hasattr(result, "to_raw"):
|
83 |
+
return result.to_raw()
|
84 |
+
if hasattr(result, "value"):
|
85 |
+
return result.value
|
86 |
+
return str(result)
|
87 |
|
88 |
# --- Basic Agent Definition ---
|
89 |
# ----- THIS IS WERE YOU CAN BUILD WHAT YOU WANT ------
|