paddlespeech.cli.vector.infer module

class paddlespeech.cli.vector.infer.VectorExecutor

Bases: BaseExecutor

Methods

disable_task_loggers()	Disable all loggers in current task.
execute(argv)	Command line entry for vector model
get_embeddings_score(enroll_embedding, ...)	get the enroll embedding and test embedding score
get_input_source(input_)	Get task input source from command line input.
infer(model_type)	Infer the model to get the embedding
postprocess()	Return the audio embedding info
preprocess(model_type, input_file)	Extract the audio feat
process_task_results(input_, results[, ...])	Handling task results and redirect stdout if needed.
show_rtf(info)	Calculate rft of current task and show results.

__call__

execute(argv: List[str]) → bool

Command line entry for vector model

Args:

argv (List[str]): command line args list

Returns:

bool:

False: some audio occurs error True: all audio process success

get_embeddings_score(enroll_embedding, test_embedding)

get the enroll embedding and test embedding score

Args:

enroll_embedding (numpy.array): shape: (emb_size), enroll audio embedding test_embedding (numpy.array): shape: (emb_size), test audio embedding

Returns:

score: the score between enroll embedding and test embedding

infer(model_type: str)

Infer the model to get the embedding

Args:

model_type (str): speaker verification model type

postprocess() → Union[str, PathLike]

Return the audio embedding info

Returns:

Union[str, os.PathLike]: audio embedding info

preprocess(model_type: str, input_file: Union[str, PathLike])

Extract the audio feat

Args:

model_type (str): speaker verification model type input_file (Union[str, os.PathLike]): audio file path