Hoi, A lot of wiktionarians are involved in recording pronunciations at the moment. At issue is that pronunciations that are recorded for scientific purposes are universally saved as wav files. Given that with the shtooka software we can record vast amounts of pronunciations, it is a waste to destroy something that can have an incredible value because of an insistence on .ogg files.
I would urge us to allow for the recording in .wav files in order to ensure that this data keeps its relevance for scientific usage.
Thanks, GerardM
Gerard,
Quoting Gerard Meijssen gerard.meijssen@gmail.com:
software we can record vast amounts of pronunciations, it is a waste to destroy something that can have an incredible value because of an insistence on .ogg files.
I would urge us to allow for the recording in .wav files in order to ensure that this data keeps its relevance for scientific usage.
Can you clarify why .wav format is more scientific that .ogg format?
Jkelly
On 2/10/07, jkelly@fas.harvard.edu jkelly@fas.harvard.edu wrote:
Gerard, Can you clarify why .wav format is more scientific that .ogg format?
I'd assume that his concern is due to Vorbis being lossy.
Considering that for a very very long time, the gold standard for much speech related research has been 8 or 16khz ulaw compressed audio, I seriously doubt Vorbis is going to impeded much research.. but, I certainly do see why we wouldn't want to just throw information away.
Oh.. *sigh* I wish we had a better way of having linked 'source' and 'distribution' forms in commons. For 99.999% of our applications, the lossily compressed file is really what we want to distribute.
On 2/10/07, Gerard Meijssen gerard.meijssen@gmail.com wrote:
Hoi, A lot of wiktionarians are involved in recording pronunciations at the moment. At issue is that pronunciations that are recorded for scientific purposes are universally saved as wav files. Given that with the shtooka software we can record vast amounts of pronunciations, it is a waste to destroy something that can have an incredible value because of an insistence on .ogg files.
I would urge us to allow for the recording in .wav files in order to ensure that this data keeps its relevance for scientific usage.
Our software will gladly allow you to upload Ogg/FLAC files (a lossless format). If we should advise the is an open question, but you are not prohibited from uploading lossless files if there is a real need.