The align
module adds timestamps to an existing speech transcript to identify the position in the audio at which each word occurred. You can use this operation to produce subtitles. This module is useful for aligning long sections of audio.
HPE recommends that you use the transcription checker directly after text normalization in your workflow. You can use the transcription checker results to make any corrections before performing alignments; for example, removing mismatched sections of audio or transcript. The correction process greatly improves the alignment results.
Mode | Input | Output | Description |
---|---|---|---|
fp
|
The name of the file to write the output to. |
Example:
w2 <- align (_, w1)
|