Update app.py

app.py

@@ -94,8 +94,7 @@ transcribe_options = dict(beam_size=3, best_of=3, without_timestamps=False)
 
 
 source_language_list = [key[0] for key in source_languages.items()]
-source_language_list_2 = [key[0] for key in DeepL_language_codes_for_translation.items()]
-translation_models_list = [key[0] for key in translation_models.items()]
+translation_models_list = [key[0] for key in DeepL_language_codes_for_translation.items()]
 
 
 device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
@@ -193,7 +192,7 @@ def speech_to_text(video_file_path, selected_source_lang, whisper_model):
         raise RuntimeError("Error Running inference with local model", e)
 
 
-def translate_transcriptions(df, selected_translation_lang_2, selected_source_lang_2):
+def translate_transcriptions(df, selected_translation_lang_2):
     if selected_translation_lang_2 is None:
             selected_translation_lang_2 = 'English'
     df.reset_index(inplace=True)
@@ -201,27 +200,27 @@ def translate_transcriptions(df, selected_translation_lang_2, selected_source_la
     print("start_translation")
     translations = []
     
-    if selected_translation_lang_2 != selected_source_lang_2:
-
-        text_combined = ""
-        for i, sentence in enumerate(df['text']):
-            if i == 0:
-                text_combined = sentence
-            else:
-                text_combined = text_combined + '\n' + sentence
-
-        data = {'text': text_combined,
-        'tag_spitting': 'xml',
-        'target_lang': DeepL_language_codes.get(selected_source_lang_2)
-               }
-        response = requests.post('https://api-free.deepl.com/v2/translate', headers=headers, data=data)
-
-        # Print the response from the server
-        translated_sentences = json.loads(response.text)
-        translated_sentences = translated_sentences['translations'][0]['text'].split('\n')
-        df['translation'] = translated_sentences
-    else:
-        df['translation'] = df['text']
+    
+
+    text_combined = ""
+    for i, sentence in enumerate(df['text']):
+        if i == 0:
+            text_combined = sentence
+        else:
+            text_combined = text_combined + '\n' + sentence
+
+    data = {'text': text_combined,
+    'tag_spitting': 'xml',
+    'target_lang': DeepL_language_codes.get(selected_source_lang_2)
+           }
+    response = requests.post('https://api-free.deepl.com/v2/translate', headers=headers, data=data)
+
+    # Print the response from the server
+    translated_sentences = json.loads(response.text)
+    translated_sentences = translated_sentences['translations'][0]['text'].split('\n')
+    df['translation'] = translated_sentences
+
+    
     print("translations done")
 
     return (df)
@@ -289,10 +288,8 @@ video_out = gr.Video(label="Video Out", mirror_webcam=False)
 
 df_init = pd.DataFrame(columns=['start','end','text'])
 df_init_2 = pd.DataFrame(columns=['start','end','text','translation'])
-selected_translation_lang = gr.Dropdown(choices=translation_models_list, type="value", value="English", label="In which language you want the transcriptions?", interactive=True)
 
 selected_source_lang = gr.Dropdown(choices=source_language_list, type="value", value="Let the model analyze", label="Spoken language in video", interactive=True)
-selected_source_lang_2 = gr.Dropdown(choices=source_language_list_2, type="value", value="English", label="Spoken language in video", interactive=True)
 selected_translation_lang_2 = gr.Dropdown(choices=translation_models_list, type="value", value="English", label="In which language you want the transcriptions?", interactive=True)
 selected_whisper_model = gr.Dropdown(choices=whisper_models, type="value", value="base", label="Selected Whisper model", interactive=True)
 
@@ -366,10 +363,9 @@ with demo:
             ##### Here you will get translated transcriptions. 
             ##### Please remember to select Spoken Language and wanted translation language
             ##### ''')
-            selected_source_lang_2.render()
             selected_translation_lang_2.render()
             translate_transcriptions_button = gr.Button("Step 3. Translate transcription")
-            translate_transcriptions_button.click(translate_transcriptions, [transcription_df, selected_translation_lang_2, selected_source_lang_2], transcription_and_translation_df)
+            translate_transcriptions_button.click(translate_transcriptions, [transcription_df, selected_translation_lang_2], transcription_and_translation_df)
             transcription_and_translation_df.render()
             
     with gr.Row():