<html><head></head><body>Just for the record, Vespucci has had various incarnations of voice input support since August 2016, currently it works as described here <a href="https://vespucci.io/help/en/17.0.0%20Release%20notes/#updated-voice-input-support-experimental">https://vespucci.io/help/en/17.0.0%20Release%20notes/#updated-voice-input-support-experimental</a><br><br>Previois versions have worked closer to what you suggest.<br><br>But you are definitely are not going to want to use it, or any other voice input system faster than at a leisurely stroll.<br><br>It isn't that it isn't technically possible, it is just that humans are dreadful at the task, and far far too slow at producing coherent voice input at any kind of useful cadence.<br><br>Simon <br><br><div class="gmail_quote">On 6 August 2022 22:10:20 CEST, Marc_marc <marc_marc@mailo.com> wrote:<blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
<pre dir="auto" class="k9mail">Hello,<br><br>Le 06.08.22 à 19:57, Ilya Zverev a écrit :<br><blockquote class="gmail_quote" style="margin: 0pt 0pt 1ex 0.8ex; border-left: 1px solid #729fcf; padding-left: 1ex;">many things are not visible on photos: e.g. streams under a highway, <br></blockquote><br>sitting in a bus, to use the previous example, I really don't see how encoding text at 50km/h would better capture the river under the road compared to a shooting: either it's not visible from the moving vehicle and photo <> text won't change anything, or you're looking in the right direction and it's visible from the bus and then typing text is often the limiting factor<br>I'll test OSMTracker tomorrow, but in principle it's impossible to say the number of floors + roof level + door location with a house every 6 meters riding with a bike, whereas it's easy to capture with a series of photos.<br><br>If this tool suits you, so much the better,<br>but I have a very different opinion from your statements<br><br><blockquote class="gmail_quote" style="margin: 0pt 0pt 1ex 0.8ex; border-left: 1px solid #729fcf; padding-left: 1ex;">Making notes solves many issues: the result is half-processed: you only need to convert text to tags<br></blockquote><br>Again, I don't see how making a note with the word "bench" that must then be converted into an osm tag would be more efficient than having<br>an application in which you write bench and which sends amenity=bench<br>to osm without any "note-to-copy-paste-into-tag" app.<br>The real improvement would be to have a voice command to control<br>an editor based on presets ("top" right bench 3-seat green wood greeen -> object created in osm, of course at 50km/h, the exact location is not easy, but it's the same issue with note) or SC (for existing objects)<br><br>Regards,<br>Marc<hr>talk mailing list<br>talk@openstreetmap.org<br><a href="https://lists.openstreetmap.org/listinfo/talk">https://lists.openstreetmap.org/listinfo/talk</a><br></pre></blockquote></div><div style='white-space: pre-wrap'><div class='k9mail-signature'>-- <br>Diese Nachricht wurde von meinem Android-Mobiltelefon mit Kaiten Mail gesendet.</div></div></body></html>