Media Data Types

This page describes all the Object types defined and used by Media activities.

Transcript

Properties

Property	Type	Description
Language	String	The language of the audio.
Text	String	The text extracted from the audio.
Segments	List<Segment>	Details segments extracted in audio transcription.

Segment

Properties

Property	Type	Description
Id	Int32	The unique identifier of the segment.
Seek	Int32	The position in the audio file where the segment begins.
Start	Double	The segment's start time in the audio file. It provides a specific timestamp indicating when the segment begins.
End	Double	The end time of the segment in the audio file. It provides a timestamp indicating when the segment ends.
Text	String	the transcribed text corresponding to the spoken content within the segment. Converts the spoken words into written text during audio transcription, and this property stores that textual representation.
Tokens	List<Int32>	Details about transcription.
Temperature	Double	The intensity or significance of the speech content within the segment, which could help prioritize or categorize segments during transcription.
AvgLogprob	Double	The average logarithmic probability associated with the transcription of the segment.
CompressionRatio	Double	The degree to which it compresses the audio data.
NoSpeechProb	Double	The probability of absence of speech within the segment.
Confidence	Double	confidence level or certainty associated with the transcription of the segment. A numerical value indicates the degree of confidence in the accuracy of the transcribed text.
Words	List<Word>	A list of individual words extracted from the transcribed text.

Word

Properties

Property	Type	Description
Text	String	The text representation of an individual word within the transcribed segment.
Start	Double	The start time of the Word within the audio segment.
End	Double	The end time of the Word within the audio segment.
Confidence	Double	the confidence level or certainty associated with the transcription of the Word. A numerical value indicates the degree of confidence in the accuracy of transcribing that specific Word.

TranscriptSearchResult

Properties

Property	Type	Description
Search Text	String	The text for the time slots were extracted.
Segments	List<FoundSegment>	The segments found for the given text.

FoundSegment

Properties

Property	Type	Description
Id	Int32	The unique identifier of the segment.
Seek	Int32	The position in the audio file where the segment begins.
Start	Double	The segment's start time in the audio file. It provides a specific timestamp indicating when the segment begins.
End	Double	The end time of the segment in the audio file. It provides a timestamp indicating when the segment ends.
Text	String	the transcribed text corresponding to the spoken content within the segment. The spoken words it converts into written text during audio transcription, and this property stores that textual representation.
Words	List<FoundWord>	A list of individual words extracted from the transcribed text.

FoundWord

Properties

Property	Type	Description
Text	String	The text representation of an individual word within the transcribed segment.
Start	Double	The start time of the Word within the audio segment.
End	Double	The end time of the Word within the audio segment.