SEARCH, improvement in ST4

Damir

Ok, i know this seems like a little thing, but it actually is a big thing when on stage and wanting to appear professional.
I am probably way ahead of times, as usual, but you never know audio segmenting a wav file did eventually occur after i asked for it when digital audio first came out😂 thanks to Peter.
Lately i have been performing by a croud that surrounds me and we have a ball, they totaly forget that i am using backing tracks as they are totally havig a great time singing and dancing , its perfect , what is the biggest problem for me while trying to make this as seamless as possible is finding the next song amongs my 450 songs.
When on stage creating a seamless performance is easy if you follow your setlist, this works OK but if you are trying to impress the audience and create the song order onthe run it is much harder then it should be , as soon as you delay, and fiddle with ST3 you break the illusion of having a full band behind you , you keep reminding them you are using backing, this could be improved so as the audience is totally engaged with you we should be able to maintain this engagement through the whole show, this would greatly improve the ST4 software.
How?
SIRI search is geat , you simply say , SIRI? who is Elvis and bang you get a result.
So why cant we just do the same type of thing to only search for our next song , the search only has 450 songs to search through , this should be a piece of cake for it to quickly either load your next song if the now song is playing or if you are in stop mode it simply finds either the exact result andgets it ready to hit start or presents all options for you to select the best option.
Surely with the technology we have now we can ask for the search capability of a particular software to have this available , make it so the search can either monitor the microphone on the device or to be able to monitor a line input that can be setup so a seperate audio input can control the search engine, so you simply dedicate a microphone out of one of the aux mixes to feed the search engine , this can be setup on stage in many ways that suit the performer and as you speak into to it , it finds the song you need immediately so you can flow through the night with ease.
No Typing needed which looks unprofessional and destroys the illusion of not having a band behind you , and even if this is not the priority , it makes life much easier for who ever is controling the show.
Surely by now this should be possible , it works great on my watch , hey siri whats the time? Siri where is Croatia? Why not SEARCH Love me tender?
Am i ahead of time again?

AnthonyB

I get the scenario and idea. In a perfect (Star Trek?) world, I can see how that would maintain that continuity of performance.

What if Siri starts babbling "I'm sorry, I didn't understand..." and you are using Headphone output and that Siri babbling is now coming out your PA?

Parking that, what you probably want is a voice search feature baked into ST4. Similar to how a Search Box where you'd type something has a small microphone icon in the right edge of it indicating you can voice search instead of typing.

I'm not an iOS developer, and then always assume Apple hamstrings you further as they are so far behind the rest of the industry, but there is an API that iOS app devs can use to leverage Siri natural language text and voice search capabilities within an app context. A very quick search seems to indicate that this is what SiriKit (https://developer.apple.com/siri/) provides (and I'm guessing with also on-ramp to Apple Intelligence whenever that finally gets launched...) but I'm going to guess there are iOS version limitations or restrictions.

Damir

AnthonyB good points you raise, but my suggestion is not to copy Siri exactly, just as we have now search does not speak back, we only need one way communication and as you say replace typing.
Example, my QC18T could be setup with one switch that mutes my Main outputs and fold back but maintains a connection to an aux out that feeds SEARCH I hit mute SPEAK IN TO SEARCH UNMUTE MY MIC and keep going with show , search brings up my results, if there is only one result it immediately loads this to be my next song , if there are multiple results you select the one you need and keep going, most songs will load automatically if you give exact name , this would eliminate any typing, and chances of audio to be heard by audience is eliminated, if you can hear your mic means you are not in search mode.This is similar to main vocalist being able to talk to stage mix engineer to tell him what he needs or to instruct backing track operator which is the next song he plans to perform .
This means one aux feed is dedicated to search and needs to go back to your device in to the search engine.
You should be able to program the word that activates/and reactivates a new search if a new choice is urgently needed.
Search result automatically exits after a short period.
A choice to use device mic or internal audio input.
Those not needing this super fast search can simply ignore it.
These sort of software facilities are what separates the Boys from Men😂.
Evens DUMB TVs😂 now days have voice search facilities , I call them dumb because they have dumb engineers creating way too complicated devices for average users.

peter

Siri can of course be integrated into apps for searching but the issue is exactly what @AnthonyB wrote. You could get a voice answer from siri blasting over the PA and I am not aware of a way to prevent that.

Damir

peter yes I understand the existing way Siri works, but that’s not what I am proposing. We don’t want Siri we simply want voice recognition that recognises our search request within the ST3 folder only and provide us with a quick and easy way to get our next song.
It does not need to respond back to us it simply needs to listen and obey. The search result cancels after a time period ready for another verbal command or it can be overwritten immediately after repeating the activation word. It does not need to be SIRI OR HEY SIRI we should be able to program in our own trigger word, just like using a finger print or face recognition, we use our voice and word choice to engage search software reacts to this and finds the closest match if it finds one result loads it as next song. Similar to the way TVs work , there is no need for audio responses, we need a simple dumb search engine that reacts to verbal command.

peter

This is way out of scope at the moment. And my personal opinion is that relying on speech recognition in a loud environment is a really bad idea.

Damir

peter This is way out of scope at the moment.

Yeah I thought so.
But it’s actually here already, it’s a pity Apple hasn’t allowed it on individual software.

Here’s how I would use it if it was available.
I can already say SIRI , what’s the time ? And it gives me the time with an included response , this is much more than I need but it appears very smart and works well.

In ST I could really use this to my advantage on stage, .
Simply activate the on board microphone while in ST ,
Now the software is listening , I provide it with trigger words that send it in to action like,
FIND SONG, PLAY , NEXT SONG ,FIND TEXT, these key words that are triggered only by my voice as I have recorded them in myself and which activate different actions ONLY IN ST and within ST folder, in other words this is a simple search engine that translates words in to text and searches / compares the text to all my songs within a playlist only if I am in playlist mode, and song mode if I am in song mode.
Example, while talking with the audience (chat, chat PAUSE! next song PAUSE ! burning love PAUSE! chat , chat, in the meantime the NEXT SONG IS LOADED AND READY TO SIMPLY HIT START WITH A 3, 4, visual count in😍😂 I think it’s doable or will be very soon.
There are plenty of apps out there successfully doing this and translating languages, so this should be possible, we don’t need a complicated SIRI or a TRANSLATOR we simply need a voice recognition that turns sound in to text and a comparator that looks at a specific area that brings up a matching result and takes appropriate action.
Like , as I am speaking to the audience and I say NEXT SONG IS , the software triggers the search enters the song title and immediately takes action to place this found file as the next song. This is the same as saying SIRI ! trigger to listen, WHATS MY SCHEDULE FOR TODAY? and whalla! we get our information with a nice humanised response which we don’t really need or want, so this is very possible already , except silly Apple don’t want us to actually make it work for us in other software.
This is very stupid of them , to me this sort of ability is worth a lot more in ST3 and on stage than it is on my watch as I can lift my hand and see the time or open my diary to see my daily schedule, but I can’t stuff around on stage looking for my next song , to me this sort of capability is worth a lot more in ST3 .
Just think if we could simply say to iPad while in ST, PLAY! love me tender, and the song starts playing , or FIND! Love, and all the LOVE results are presented to choose from or NEXT SONG ! Love me tender, and it gets loaded as the next song, this would eliminate a lot of unnecessary typing and ugliness on stage, I can already do that on my google music player devices I have around the home , I simply say Hey Google play BURNING LOVE, the device connects wirelessly to my router searches the whole world and finds BURNING LOVE and I am listening to it within seconds.
This is already here, so why don’t we have it in dedicated software in a much more simpler way.
We don’t need responses , we don’t need to search the world , we only need to search a small specific area on the device and take very limited action once the file is found. THIS SHOULD BE AVAILABLE TO US ALREADY,
I had powered speakers at least 10 years before they came out in the market not because I was a genius , because I came up with the idea and took action, this applies to this idea, someone is going to do this it’s just a matter of it being done.
As I said, language translators are already common, voice search is already common it’s a matter of putting these capabilities to good use.
I thought AI was already here, where is it? I don’t see it?😂

To add, we have bluetooth ear mics that phones use, why not use that as the sound sourcre to the audio input to ST that is used for verbal remote contr, we normally use audio output for playing music, why not make good use of the audio input to ST for a very useful purpose, this could also be used to jump audio segments verbally , X to Verse, X to Chorus 1 , X to Solo 2, X to End, X TO is the trigger command and Verse, Chorus 1, Solo 2 are the audio segment names , as you say X to triggers the listen mode the name has to match the exact writing of the segment , if theres a match it selects that segment as NEXT TO PLAY , if you change your mind say X Cancel , or X End if you want to jump to ending instead at the last moment.
You simply mute your vocal mic and give verbal command or move away from the main mic and speak in to bluetooth in ear mic, which is constantly monitoring for correct command to do something this should be very possible if given enough thought, even if the song is playing or not you sould be able to control audio segments and search song functions , no fuss no foot dancing or touching the screen, AI is here so lets use it.
All these ideas are just thrown together without much thought of naming , but its the overall conceipt that i am trying to point out , its so nice to verbally say hey Google! Play Croatian top hits and presto it just works, so why cant we do this with individual software?