Not Getting Correct Ouptut / Getting Same Ouput everytime

[Himalmevada]

I have done all the step to perform asr on DOGRI Language.
Everytime i give new audio it gives the same output like this:

OUTPUT :
Traceback (most recent call last):
File "/home/himal/Documents/Vakyansh_AI_MODEL/scripts/inference/../../utils/inference/single_file_inference.py", line 380, in
result = parse_transcription(args_local.model, args_local.dict, args_local.wav, args_local.cuda, args_local.decoder, args_local.lexicon, args_local.lm_path, args_local.half)
File "/home/himal/Documents/Vakyansh_AI_MODEL/scripts/inference/../../utils/inference/single_file_inference.py", line 349, in parse_transcription
generator = W2lKenLMDecoder(args, target_dict)
File "/home/himal/Documents/Vakyansh_AI_MODEL/scripts/inference/../../utils/inference/single_file_inference.py", line 188, in init
assert (
AssertionError: ['ळ', 'न', 'प', 'क', 'म', '|'] [3, 39, 40, 20, 44, 4]