import json
import os
from dataclasses import asdict
from pprint import pprint

from sziszapangma.integration.service_core.asr.asr_result import AsrResult

RESPONSE_DIR = 'asr_responses_cache/common_voice_wav2vec2'
RESULT_DIR = 'experiment_data/cached_asr/common_voice_wav2vec2'

if __name__ == '__main__':
    for file_name in os.listdir(RESPONSE_DIR):
        print(file_name)
        d = json.load(open(f'{RESPONSE_DIR}/{file_name}', 'r'))
        if True:
            # all_text = ' '.join([result['alternatives'][0]['transcript'] for result in d['results']
            #                      if 'transcript' in result['alternatives'][0]])
            # all_words = []
            # for result in d['results']:
            #     if 'transcript' in result['alternatives'][0]:
            #         all_words.extend([it['word'] for it in result['alternatives'][0]['words']])
            # r = {
            #     'words_time_alignment': None,
            #     'transcription': all_words,
            #     'full_text': all_text
            # }
            json.dump(d, open(f'{RESULT_DIR}/{file_name}', 'w'))
        else:
            print(d)
            print('err')