Data structures — Mozilla DeepSpeech 0.9.3 documentation (original) (raw)

Metadata¶

struct Metadata¶

An array of CandidateTranscript objects computed by the model.

Public Members

const CandidateTranscript *const transcripts¶

Array of CandidateTranscript objects

const unsigned int num_transcripts¶

Size of the transcripts array

CandidateTranscript¶

struct CandidateTranscript¶

A single transcript computed by the model, including a confidence value and the metadata for its constituent tokens.

Public Members

const TokenMetadata *const tokens¶

Array of TokenMetadata objects

const unsigned int num_tokens¶

Size of the tokens array

const double confidence¶

Approximated confidence value for this transcript. This is roughly the sum of the acoustic model logit values for each timestep/character that contributed to the creation of this transcript.

TokenMetadata¶

struct TokenMetadata¶

Stores text of an individual token, along with its timing information.

Public Members

const char *const text¶

The text corresponding to this token

const unsigned int timestep¶

Position of the token in units of 20ms

const float start_time¶

Position of the token in seconds