Update modules/translation/translation_base.py
Browse files
modules/translation/translation_base.py
CHANGED
|
@@ -129,6 +129,56 @@ class TranslationBase(ABC):
|
|
| 129 |
finally:
|
| 130 |
self.release_cuda_memory()
|
| 131 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 132 |
def offload(self):
|
| 133 |
"""Offload the model and free up the memory"""
|
| 134 |
if self.model is not None:
|
|
|
|
| 129 |
finally:
|
| 130 |
self.release_cuda_memory()
|
| 131 |
|
| 132 |
+
def translate_text(self,
|
| 133 |
+
input_list_dict: list,
|
| 134 |
+
model_size: str,
|
| 135 |
+
src_lang: str,
|
| 136 |
+
tgt_lang: str,
|
| 137 |
+
max_length: int = 200,
|
| 138 |
+
add_timestamp: bool = True,
|
| 139 |
+
progress=gr.Progress()) -> list:
|
| 140 |
+
"""
|
| 141 |
+
Translate text from source language to target language
|
| 142 |
+
Parameters
|
| 143 |
+
----------
|
| 144 |
+
str_text: str
|
| 145 |
+
List[dict] to translate
|
| 146 |
+
model_size: str
|
| 147 |
+
Whisper model size from gr.Dropdown()
|
| 148 |
+
src_lang: str
|
| 149 |
+
Source language of the file to translate from gr.Dropdown()
|
| 150 |
+
tgt_lang: str
|
| 151 |
+
Target language of the file to translate from gr.Dropdown()
|
| 152 |
+
max_length: int
|
| 153 |
+
Max length per line to translate
|
| 154 |
+
add_timestamp: bool
|
| 155 |
+
Boolean value that determines whether to add a timestamp
|
| 156 |
+
progress: gr.Progress
|
| 157 |
+
Indicator to show progress directly in gradio.
|
| 158 |
+
I use a forked version of whisper for this. To see more info : https://github.com/jhj0517/jhj0517-whisper/tree/add-progress-callback
|
| 159 |
+
Returns
|
| 160 |
+
----------
|
| 161 |
+
A List of
|
| 162 |
+
List[dict] with translation
|
| 163 |
+
"""
|
| 164 |
+
try:
|
| 165 |
+
self.cache_parameters(model_size=model_size,src_lang=src_lang,tgt_lang=tgt_lang,max_length=max_length,add_timestamp=add_timestamp)
|
| 166 |
+
self.update_model(model_size=model_size,src_lang=src_lang,tgt_lang=tgt_lang,progress=progress)
|
| 167 |
+
|
| 168 |
+
total_progress = len(input_list_dict)
|
| 169 |
+
for index, dic in enumerate(input_list_dict):
|
| 170 |
+
progress(index / total_progress, desc="Translating..")
|
| 171 |
+
translated_text = self.translate(dic["text"], max_length=max_length)
|
| 172 |
+
dic["text"] = translated_text
|
| 173 |
+
|
| 174 |
+
return input_list_dict
|
| 175 |
+
|
| 176 |
+
except Exception as e:
|
| 177 |
+
print(f"Error translating file: {e}")
|
| 178 |
+
raise
|
| 179 |
+
finally:
|
| 180 |
+
self.release_cuda_memory()
|
| 181 |
+
|
| 182 |
def offload(self):
|
| 183 |
"""Offload the model and free up the memory"""
|
| 184 |
if self.model is not None:
|