KeyError 'en'

[pr-data-port]

I am getting this error when trying to align english model with whisperx:

KeyError                                  Traceback (most recent call last)
Cell In[22], line 1
----> 1 result_aligned = whisperx.align(
      2     result["segments"], 
      3     model_a, 
      4     metadata, 
      5     audio_resampled, 
      6     device, 
      7     return_char_alignments=False)

File /data/anaconda3/envs/vsep/lib/python3.9/site-packages/whisperx/alignment.py:129, in align(transcript, model, align_model_metadata, audio, device, interpolate_method, return_char_alignments, print_progress, combined_progress, preprocess)
    126 # Load align model Huggingface processor for audio feature extraction (Normalization)
    127 if preprocess and model_type == 'huggingface':
    128     processor = Wav2Vec2Processor.from_pretrained(
--> 129         DEFAULT_ALIGN_MODELS_HF[model_lang])
    131 # 1. Preprocess to keep only characters in dictionary
    132 total_segments = len(transcript)

KeyError: 'en'

the whisperx align model I use:

model_a, metadata = whisperx.load_align_model(
    model_name='jonatasgrosman/wav2vec2-large-xlsr-53-english', 
    language_code=result["language"], 
    device=device)

result_aligned = whisperx.align(
    result["segments"], 
    model_a, 
    metadata, 
    audio_resampled, 
    device, 
    return_char_alignments=False)    

has anyone had any similar issue?