Update app.py

app.py

@@ -16,16 +16,51 @@ def process_audio_file(file):
     
 def transcribe(target_language, file):
     
-    print("Target", target_language)
+    target_code = target_language.split("(")[-1].split(")")[0]
+    forced_bos_token_id = MAPPING[target_code]
 
     input_values = process_audio_file(file)
     
-    sequences = model.generate(input_values)
+    sequences = model.generate(input_values, forced_bos_token_id=forced_bos_token_id)
     
     transcription = tokenizer.batch_decode(sequences, skip_special_tokens=True)
     return transcription[0]
     
-target_languages = ["German", "French", "Italian"]
+target_languages = [
+    "German (de)",
+    "Turkish (tr)",
+    "Persian (fa)",
+    "Swedish (sv)",
+    "Mongolian (mn)",
+    "Chinese (zh)",
+    "Welsh (cy)",
+    "Catalan (ca)",
+    "Slovenian (sl)",
+    "Estonian (et)",
+    "Indonesian (id)",
+    "Arabic (ar)",
+    "Tamil (ta)",
+    "Latvian (lv)",
+    "Japanese (ja)",
+]
+
+MAPPING = {
+    "de": 250003,
+    "tr": 250023,
+    "fa": 250029,
+    "sv": 250042,
+    "mn": 250037,
+    "zh": 250025,
+    "cy": 250007,
+    "ca": 250005,
+    "sl": 250052,
+    "et": 250006,
+    "id": 250032,
+    "ar": 250001,
+    "ta": 250044,
+    "lv": 250017,
+    "ja": 250012,
+}
     
 iface = gr.Interface(
     fn=transcribe,