diff --git a/subgen.py b/subgen.py
index ae021fb..38146be 100644
--- a/subgen.py
+++ b/subgen.py
@@ -24,6 +24,7 @@ import av
 import ffmpeg
 import whisper
 import re
+import ast
 from watchdog.observers.polling import PollingObserver as Observer
 from watchdog.events import FileSystemEventHandler
 import faster_whisper
@@ -102,6 +103,7 @@ def update_env_variables():
     custom_regroup = os.getenv('CUSTOM_REGROUP', 'cm_sl=84_sl=42++++++1')
     detect_language_length = os.getenv('DETECT_LANGUAGE_LENGTH', 30)
     skipifexternalsub = convert_to_bool(os.getenv('SKIPIFEXTERNALSUB', False))
+    kwargs = ast.literal_eval(os.getenv('SUBGEN_KWARGS', ''))
 
     set_env_variables('subgen.env')
     
@@ -444,9 +446,9 @@ async def asr(
         if model_prompt:
             custom_prompt = greetings_translations.get(language, '') or custom_model_prompt
         if custom_regroup:
-            result = model.transcribe_stable(audio_data, task=task, input_sr=16000, language=language, progress_callback=progress, initial_prompt=custom_prompt, regroup=custom_regroup)
+            result = model.transcribe_stable(audio_data, task=task, input_sr=16000, language=language, progress_callback=progress, initial_prompt=custom_prompt, regroup=custom_regroup, **kwargs)
         else:
-            result = model.transcribe_stable(audio_data, task=task, input_sr=16000, language=language, progress_callback=progress, initial_prompt=custom_prompt)
+            result = model.transcribe_stable(audio_data, task=task, input_sr=16000, language=language, progress_callback=progress, initial_prompt=custom_prompt, **kwargs)
         appendLine(result)
         elapsed_time = time.time() - start_time
         minutes, seconds = divmod(int(elapsed_time), 60)
@@ -485,7 +487,7 @@ async def detect_language(
         task_queue.put(task_id)
         
         audio_data = np.frombuffer(audio_file.file.read(), np.int16).flatten().astype(np.float32) / 32768.0
-        detected_language = model.transcribe_stable(whisper.pad_or_trim(audio_data, int(detect_language_length) * 16000), input_sr=16000).language
+        detected_language = model.transcribe_stable(whisper.pad_or_trim(audio_data, int(detect_language_length) * 16000), input_sr=16000, **kwargs).language
         # reverse lookup of language -> code, ex: "english" -> "en", "nynorsk" -> "nn", ...
         language_code = get_key_by_value(whisper_languages, detected_language)
 
@@ -546,10 +548,10 @@ def gen_subtitles(file_path: str, transcription_type: str, force_language=None)
         if custom_regroup:
             result = model.transcribe_stable(file_path, language=force_language, task=transcription_type,
                                              progress_callback=progress, initial_prompt=custom_model_prompt,
-                                             regroup=custom_regroup)
+                                             regroup=custom_regroup, **kwargs)
         else:
             result = model.transcribe_stable(file_path, language=force_language, task=transcription_type,
-                                             progress_callback=progress, initial_prompt=custom_model_prompt)
+                                             progress_callback=progress, initial_prompt=custom_model_prompt, **kwargs)
 
         appendLine(result)
         file_name, file_extension = os.path.splitext(file_path)