-
-
Notifications
You must be signed in to change notification settings - Fork 141
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Sometimes a song will have some audio of the last song in the beginning #359
Comments
Not sure if this is related, but with Spotify Free some songs contain 1s of the previous advertisment at the beginning. However, if there was no advertisment before, they start on point for me. |
Looking at the audio waves in audacity, it looks like the bit of the last song ending up in the next song is always lasting exactly 1s. I'm not sure where I could start looking to fix the bug (I'm not much of a C# guy), but could some delay somewhere in the code could be causing this? |
So, after digging a bit in the code, my theory is that when the The "1s" number I referred to in my previous comment seems irrelevant, it'll depend on the conditions the program is running in (i.e. how long the CPU takes to finish writing bytes read from My workaround for now is to add a delay of about ~800ms (again, determined for my machine) before triggering the I guess one way to solve this properly could be to detect a silence in the |
@bziorkiewicz has funded $25.00 to this issue.
|
A short-term workaround could be an option in the settings to delay next recording by X ms. I tried to just edit these mp3 files - cut from the next one and paste into the previous but there seems to be a very short gap between recordings, so even after merge there is a very short silence in the place of merging. |
It looks like there is no easy way to fix this automatically. I'll propose another approach. This bug happens when there is no silence at the end of a song. After such a song, the next played song has an unwanted sound prefix. So, such songs without a silence suffix at the end must be handled manually. If only SpyTify could detect such songs and skip recording the next song after it (because it would have a buggy prefix) and print this in the logs, user then would manually replay those unrecorded songs and record them; this time working because problematic songs (without silence suffix) are not in the playlist any more. |
I noticed that @bziorkiewicz has suggested a good idea in another issue (closed): controlling playing from directly SpyTify which would allow waiting a few seconds between each song. |
I'm not very familiar with GitHub, but I see this marked as "Done" and progress of the next release as 80%. This is a really needed feature. Is there any funding that would speed that up? |
As @guillaumealgis told on previous comments, the MainWindowTitle of Spotify process changes the title of the track that is played to the next song one second before the actual track is ended, this happends on all tracks, but it's noticed only on tracks that not finish with a silence, like @maliayas pointed. I think the proposal of @guillaumealgis is great, is a good approach because personally I don't mind if a song contains 1s of silence at the end of the song, as long as the song not contains audio from the previous track at the start, or ends up with a cut in the song. |
This is stil a work in progress, I even created a fake spotify inside the project to test the AudioThrottler that it's still not working properly. I cannot work on it as often as I want, since I'm on a Mac now. and don't have access at all time to a windows pc. Even if this throttler does work, I don't know how it will be able to split two tracks that have no silence in between. Adding just a delay was not enough, Spotify changes its title randomly, so sometimes the delay needed was -200ms, sometimes it was 400ms or 800ms, and some other times no delay. This is kind of tricky. |
At least previously (think v1.11.1) it would sometimes also not correctly split for me, if there was a silence between the songs. |
Hello, my solution to this problem was to edit into project EspionSpotify.sln: |
i looked around on this issue, and wanted to clarify if everyone here didn't have the |
I think that's not the problem for me, as it's a custom playlist and crossfade is off. |
I don't know if progress was made in this issue but what I do manually is restart the track as a one-off record. Seamless playback while ongoing recording introduces some artifacts. For songs that begin with the tail end of previous song or an ad clip I would get much better gap transitions between songs by doing this:
Repeat this process for any of the affected songs. It's a bit tedious but this is one way to replace tracks. Does the S-API allow for some client control in tandem with Spytify running in the background? Perhaps even sending a hot key is a workaround. |
The AudioThrottler now works, it will be available in the next release |
when do you think the new version will be released? |
is there a branch we can compile to try this out? |
if jwallet is still around i'm even willing to put a few more beers on his desk for this resolution. I can't tell if this project is abandoned, on hiatus or just really really stuck on a problem. Speak up if you're out there my guy, hope you're well! |
Uploading copy_6E5D6BB6-9FBD-4784-B2E9-84DE3F7B4FAA.mov… |
Most songs will have extra leading silence. Some songs will have a bit of audio from the last track in them.
I made this script and after running it I'm happy with how the songs begin. Some songs might be trunkated! But I prefer this over the silence/junk:
Is there some "feature" that I have accidentally turned on that causes this?
(I looked through old issues and couldn't find this mentioned, sorry in advance if I missed it.)
IssueHunt Summary
Backers (Total: $25.00)
Become a backer now!
Or submit a pull request to get the deposits!
Tips
The text was updated successfully, but these errors were encountered: