Open Audacity by double-clicking its desktop icon. Once it has loaded, click on the "File" menu and select "Open." A file browser dialogue will appear. Locate the audio file containing the vocals you want to stretch and double-click it. Audacity will load and display the file's waveform on a new track.
Click "Play" to play back the vocal track. Watch the play header move over the waveform as it plays. The peaks in the waveform correspond to the loudest parts of the track; the valleys correspond to the quietest parts of the track. As a result, it is possible to determine where individual words occur in the waveform. For example, if the vocalist sings "Mary had a little lamb," the first syllable of "Mary" will be the first peak in the waveform. The waveform will decline somewhat at the end of each syllable and rise again on the next. Because singers tend to string their words together, the waveform will rarely decline all the way to zero decibels (dBs), but the individual syllables and words will still be distinguishable as waveform peaks.
Click on the "Selection Tool" button in the toolbar. The icon on the button resembles a standard insertion cursor as seen in most word processors and text editors. This cursor is used to select specific sections of an overall track.
Click on the the vocal track just before the first waveform peak. Ideally, the point at which you click should be at or near zero dBs. This is the beginning of the first syllable or word in the vocal track. When you click, Audacity will create a "split point" at that location.
In the "Edit" menu, select "Split." The vocal track will be split at the split point. Click on the right-hand end of the first waveform peak. This represents the end of the first syllable or word in the recording. Select "Split" from the "Edit" menu again. The track will be split at that point.
Repeat steps four and five for each peak in the waveform. The goal is to split each individual syllable or word into its own audio region, which can then be manipulated as desired. The "stretching" effect is achieved by click-dragging the individual audio regions to whatever location in the timeline that you want them to occur. As a result, it is possible to put as much time between each syllable or word as you like, without altering the pitch of the vocals.