Transcribing Documents
You can use Reveal to transcribe audio and video files into linked searchable text.
Once you have identified and selected the files for transcription, open Transcribe from the Review Toolbar above the Document List.
 |
Once you have chosen the documents to be processed, set the following:
The Text Set in which the original documents are to be found (usually Native / HTML).
The language in use from the following list of languages and dialects:
Arabic
Chinese (Simplified)
Chinese (Traditional)
Czech
Danish
Dutch
English
Finnish
French
US English
Welsh English
Spanish
US Spanish
French
Canadian French
Farsi
German
Swiss German
Hebrew
Indian Hindi
Indonesian
Italian
Japanese
Korean
Malay
Portuguese
Brazilian Portuguese
Russian
Tamil
Telugu
Turkish
The Text Set into which the transcription should be placed (usually Transcription).
Note
Searching Transcription Text - The Text Set containing the Transcription text must be indexed before it can be searched.
A Job Name should be supplied to track the progress and result under Jobs > AV Transcription under the Flyout Menu.
You may reference the underlying technical detail for transcription at https://docs.aws.amazon.com/transcribe/latest/dg/input.html . Within its guidance on containers and formats for transcription you are advised:
When you transcribe an audio file or video file using the StartTranscriptionJob API or the Amazon Transcribe console, make sure that the file is:
In FLAC, MP3, MP4, Ogg, WebM, AMR, or WAV file format.
Less than 4 hours in length and less than 2 GB in size (500 MB for call analytics jobs).
Note
For AMR, Amazon Transcribe supports both Adaptive Multi-Rate Wideband (AMR-WB) and Adaptive Multi-Rate Narrowband (AMR-NB) codecs.
For the Ogg and WebM file formats, Amazon Transcribe supports the Opus codec.
Note
Although Reveal Review can transcribe FLAC and AMR file types, they are not currently supported by our in-browser media player.
For best results:
Use a lossless format. You can choose either FLAC, or WAV with PCM 16-bit encoding.
Use a sample rate of 8,000 Hz for telephone audio.