asr_datsets = {'LibriSpeech-Test-Clean': 'aa', 
                'LibriSpeech-Test-Other': 'bb', 
                'Common-Voice-15-En-Test': 'cc', 
                'Peoples-Speech-Test': 'dd', 
                'GigaSpeech-Test': 'ee', 
                'Earnings21-Test': 'ff', 
                'Earnings22-Test': 'gg', 
                'Tedlium3-Test': 'hh', 
                'Tedlium3-Longform-Test': 'ii', 
                'IMDA-Part1-ASR-Test': 'jj', 
                'IMDA-Part2-ASR-Test': 'kk', 
                'IMDA-Part3-ASR-Test': 'll', 
                'IMDA-Part4-ASR-Test': 'mm', 
                'IMDA-Part5-ASR-Test': 'nn', 
                'IMDA-Part6-ASR-Test': 'oo'
                }

sqa_datasets = {'CN-College-Listen-MCQ-Test': 'aa', 
                'DREAM-TTS-MCQ-Test': 'bb', 
                'SLUE-P2-SQA5-Test': 'cc', 
                'Public-SG-Speech-QA-Test': 'dd', 
                'Spoken-Squad-v1': 'ee'
                }

si_datasets = {'OpenHermes-Audio-Test': 'aa', 
               'ALPACA-Audio-Test': 'bb'
               }

ac_datasets = {
    'WavCaps-Test': 'aa', 
    'AudioCaps-Test': 'bb'
}

asqa_datasets = {
    'Clotho-AQA-Test': 'aa', 
    'WavCaps-QA-Test': 'bb', 
    'AudioCaps-QA-Test': 'cc'
}

er_datasets = {
    'IEMOCAP-Emotion-Test': 'aa', 
    'MELD-Sentiment-Test': 'bb', 
    'MELD-Emotion-Test': 'cc'
}

ar_datsets = {
    'VoxCeleb-Accent-Test': 'aa'
}

gr_datasets = {
    'VoxCeleb-Gender-Test': 'aa', 
    'IEMOCAP-Gender-Test': 'bb'
}

spt_datasets = {
    'Covost2-EN-ID-test': 'aa', 
    'Covost2-EN-ZH-test': 'bb',
    'Covost2-EN-TA-test': 'cc', 
    'Covost2-ID-EN-test': 'dd', 
    'Covost2-ZH-EN-test': 'ee', 
    'Covost2-TA-EN-test': 'ff'
}

cnasr_datasets = {
    'Aishell-ASR-ZH-Test': 'aa'
}

metrics = {
    'wer': '11',
    'llama3_70b_judge_binary': '22',
    'llama3_70b_judge': '33',
    'meteor': '44',
    'bleu': '55'
}