Similar to the proposal in #59, it might a nice feature to listen to several files from a dataset.
But this includes several challenges:
How to provide an interface to select an audio file. If the dataset contains 1,000,000 files a dropdown menu does not sound like a good idea. Maybe one can provide an option to provide a table name, or a scheme + certain value, and it then selects a random sample matching those selections
We cannot pre-fetch all possible files and store them with the HTML page, so we need a process that is able to load the files on the fly when requested. This might still be risky as some datasets have bundled single files and huge archives containing several files
Instead of the interactive approach, we might also come up with a mechanism that provides more than one example on the static page, by selecting randomly 10 samples.
For now, I would recommend very low priority for this issue.
Similar to the proposal in #59, it might a nice feature to listen to several files from a dataset.
But this includes several challenges:
Instead of the interactive approach, we might also come up with a mechanism that provides more than one example on the static page, by selecting randomly 10 samples.
For now, I would recommend very low priority for this issue.