from experiment.duckling.duckling_asr_processing_task import DucklingAsrProcessingTask from experiment.duckling.duckling_gold_transcript_processing_task import DucklingGoldTranscriptProcessingTask from experiment.pos_processing.asr_spacy_token_pos_processing_task import AsrSpacyTokenPosProcessingTask from experiment.pos_processing.ner_spacy_wer_processing_task import NerSpacyWerProcessingTask from experiment.pos_processing.spacy_pos_wer_processing_task import SpacyPosWerProcessingTask from experiment.pos_processing.tag_spacy_wer_processing_task import TagSpacyWerProcessingTask from experiment.voicelab.voicelab_dependency import get_record_provider, get_repository, GOLD_TRANSCRIPT, \ GOLD_TRANSCRIPT_SPACY, TECHMO_POLISH_ASR, WORD_TECHMO_METRICS_WER, WORD_TECHMO_ALIGNMENT_WER, TECHMO_SPACY, \ POS_TECHMO_METRICS_WER, POS_TECHMO_ALIGNMENT_WER, WORD_TECHMO_METRICS_WER_EMBEDDINGS, \ WORD_TECHMO_ALIGNMENT_WER_EMBEDDINGS, TAG_SPACY_TECHMO_ALIGNMENT_WER_EMBEDDINGS, \ TAG_SPACY_TECHMO_METRICS_WER_EMBEDDINGS, AJN_SPACY, TAG_SPACY_AJN_ALIGNMENT_WER_EMBEDDINGS, \ TAG_SPACY_AJN_METRICS_WER_EMBEDDINGS, NER_SPACY_TECHMO_METRICS_WER_EMBEDDINGS, \ NER_SPACY_TECHMO_ALIGNMENT_WER_EMBEDDINGS, AJN_POLISH_ASR, NER_SPACY_AJN_ALIGNMENT_WER_EMBEDDINGS, \ NER_SPACY_AJN_METRICS_WER_EMBEDDINGS from sziszapangma.core.transformer.web_embedding_transformer import WebEmbeddingTransformer from sziszapangma.integration.asr_processor import AsrPathCacheClient from sziszapangma.integration.experiment_manager import ExperimentManager from sziszapangma.integration.task.asr_task import AsrTask from sziszapangma.integration.task.classic_wer_metric_task import ClassicWerMetricTask from sziszapangma.integration.task.embedding_wer_metrics_task import EmbeddingWerMetricsTask def run_voicelab_experiment(): record_provider = get_record_provider() experiment_processor = ExperimentManager( record_id_iterator=record_provider, processing_tasks=[ DucklingGoldTranscriptProcessingTask( task_name='DucklingGoldTranscriptProcessingTask', require_update=False, duckling_endpoint='http://localhost:8080/parse', duckling_language='pl_PL', input_property_name=GOLD_TRANSCRIPT, duckling_property_name='gold_transcript_duckling' ), DucklingAsrProcessingTask( task_name='techmo_DucklingGoldTranscriptProcessingTask', require_update=False, duckling_endpoint='http://localhost:8080/parse', duckling_language='pl_PL', input_property_name=TECHMO_POLISH_ASR, duckling_property_name='techmo_duckling' ), DucklingAsrProcessingTask( task_name='ajn_DucklingGoldTranscriptProcessingTask', require_update=False, duckling_endpoint='http://localhost:8080/parse', duckling_language='pl_PL', input_property_name=AJN_POLISH_ASR, duckling_property_name='ajn_duckling' ) ], experiment_repository=get_repository(), relation_manager_provider=record_provider ) experiment_processor.process() def example_run(): run_voicelab_experiment() if __name__ == '__main__': example_run()