Here's how I imported a recording of a 1000-item word list into Flex. As I remember, I started out by typing the data in Excel (or LibreOffice Calc). I labeled the column headers with the SFM fields which correspond to the Flex fields in which I want my data to end up and typed in all of my data. If I recall correctly, SheetSwiper needs these SFM labels in a later step.
- \rf -- Reference (I used speaker name+sequence number, e.g. "Jon001".)
- \ge -- English gloss
- \lx -- Lexeme
- \nt -- Note (this field was blank for most entries)
- \so -- Source (the date on which I elicited the data. The speaker's name is in the \rf field.)
- \pf -- Pronunciation file
- Split up the long recordings in Audacity. Use Track...Add Label at Selection (Ctrl-B) to mark the beginning of each utterance. Delete any extraneous conversation, noises, etc, then use File...Export Multiple to save each utterance in a separate WAV file named according to the label you assigned using Add Label. Make the labels and file names match your \rf field.
- If you want the WAV files to be included in Send-Receive, then copy them all to C:\ProgramData\SIL\FieldWorks\Projects\projectname\LinkedFiles\AudioVisual. If you don't want the .WAV files included in Send-Receive, then you should save them in a different folder.
- Make sure the \so column in your Excel file matches the location of your WAV files. Use Excel's CONCATENATE function (=CONCATENATE("C:\Progr...\Jon",TEXT(E2,"000"),".wav")) to generate the \pf column. Type a couple entries in \rf, then select them and drag the bottom right corner of the selection box to fill the rest of the column all at once.
- Use SheetSwiper to convert the Excel file to SFM.
- Do the Flex import. Import the WAV file name into the Pronunciation (Media File) field. I did NOT use an audio writing system, but I do have a play button next to my filenames in Flex.