Output timestamps for start of each track?

[thatpaulschofield]

Is your feature request related to a problem? Please describe.
I'm using py-crossfade as part of a workflow to create music video compilations. After generating the crossfaded audio with py-crossfade, I bring it into DaVinci Resolve and lay the video clips on top of the generated audio, syncing the video to the audio by hand. It occurred to me that with a little metadata, I could automate the last part.

Describe the solution you'd like
A log file with timestamps of where each track starts within the crossfaded track generated by py-crossfade. I figure I could use this data to call ffmpeg to generate the compiled video automatically.

Thank you for the time and effort to create this wonderful software! 🙏🙏🙏

[oguzhan-yilmaz]

Hey @thatpaulschofield, it's great that this piece of software is used by people, and thanks for your nice words.

I've been putting off working on this project for the longest time, and your comment had me roll up my sleeves.

I've noticed that every dependency went out of date, and it's hell to get pyCrossfade working. So I've created a Dockerfile to nail all these values down.

And scripted usage didn't feel solid, so I've created a CLI for easier interaction.

Please try the new docker container CLI and with the pycrossfade crossfade --verbose option, and tell me what you think and what you'd need?

(I've only implemented the --verbose flag for crossfade and not crossfade-many, I'll be updating it based on your feedback)

[thatpaulschofield]

Thanks so much for your attention! I wanted to reply while I've got a second and let you know that I am going to try the docker file. This is a big win for user experience. So a little background: I'm currently using Windows, but accustomed to using Windows Subsystem for Linux quite frequently with these sorts of tools. I'm currently using an ugly mix of DOS batch scripts, Powershell scripts and Python to download the videos from Youtube, extract the audio, use demucs to remove the bass track, and then run the pyCrossfade to stitch the audio together (before manually aligning the videos to the crossfaded audio in Davinci Resolve). I'll try the --verbose option and see if that gives enough information to automate the video editing part. I am currently running it on at least a dozen audio files at a time, so the support for crossfade-many would be a godsend. Actually, it's possible that the song option will do what I need. I'll take a look at that as well. I have one unusual situation that I did find a workaround for by editing the original audio, but let me tell you about it. The song was "Attention" by NewJeans. The intro is completely syncopated, which throws off the downbeat detection. I just edited out that intro section and it worked fine. I wonder if there's a way to give the downbeat detection a hint about that kind of scenario, or does it just make more sense to remove any syncopated intros from a song? Btw, it would be silly to ask you to support the dependencies you use, and I won't do that. However, do you have any advice for what to do when they throw an obscure error? I have a couple of different songs that I just haven't been able to use so far. I'll hit up whoever manages those libraries, of course! If it's useful to you, I can send you some of the error messages. I plan on getting back to you in a few days. Thanks again for the great software!

…

On Mon, Dec 9, 2024 at 12:46 PM Oguzhan Yilmaz ***@***.***> wrote: Hey @thatpaulschofield <https://github.com/thatpaulschofield>, it's great that this piece of software is used by people, and thanks for your nice words. I've been putting off working on this project for the longest time, and your comment had me roll up my sleeves. I've noticed that every dependency went out of date, and it's hell to get pyCrossfade working. So I've created a Dockerfile to nail all these values down. And scripted usage didn't feel solid, so I've created a CLI for easier interaction. Please try the new docker container CLI and with the pycrossfade crossfade --verbose option, and tell me what you think and what you'd need? (I've only implemented the --verbose flag for crossfade and not crossfade-many, I'll be updating it based on your feedback) — Reply to this email directly, view it on GitHub <#8 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AACCIQV4DKR7KBP7VACZE4T2EXQQJAVCNFSM6AAAAABSUIYSBKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDKMRZGA3TQNRTGI> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[thatpaulschofield]

Hi there! I've been playing around with the new Docker version. I was wondering what it means when you run pycrossfade crossfade or pycrossfade crossfade-many, and it just hangs indefinitely? Like overnight? I'm running crossfade with the -v switch and I don't see anything output to the console. Is there a way to get to the logs inside the container in cases where it hangs? Thanks! On Wed, Dec 11, 2024 at 2:18 PM Paul Schofield ***@***.***> wrote:

…

Thanks so much for your attention! I wanted to reply while I've got a second and let you know that I am going to try the docker file. This is a big win for user experience. So a little background: I'm currently using Windows, but accustomed to using Windows Subsystem for Linux quite frequently with these sorts of tools. I'm currently using an ugly mix of DOS batch scripts, Powershell scripts and Python to download the videos from Youtube, extract the audio, use demucs to remove the bass track, and then run the pyCrossfade to stitch the audio together (before manually aligning the videos to the crossfaded audio in Davinci Resolve). I'll try the --verbose option and see if that gives enough information to automate the video editing part. I am currently running it on at least a dozen audio files at a time, so the support for crossfade-many would be a godsend. Actually, it's possible that the song option will do what I need. I'll take a look at that as well. I have one unusual situation that I did find a workaround for by editing the original audio, but let me tell you about it. The song was "Attention" by NewJeans. The intro is completely syncopated, which throws off the downbeat detection. I just edited out that intro section and it worked fine. I wonder if there's a way to give the downbeat detection a hint about that kind of scenario, or does it just make more sense to remove any syncopated intros from a song? Btw, it would be silly to ask you to support the dependencies you use, and I won't do that. However, do you have any advice for what to do when they throw an obscure error? I have a couple of different songs that I just haven't been able to use so far. I'll hit up whoever manages those libraries, of course! If it's useful to you, I can send you some of the error messages. I plan on getting back to you in a few days. Thanks again for the great software! On Mon, Dec 9, 2024 at 12:46 PM Oguzhan Yilmaz ***@***.***> wrote: > Hey @thatpaulschofield <https://github.com/thatpaulschofield>, it's > great that this piece of software is used by people, and thanks for your > nice words. > > I've been putting off working on this project for the longest time, and > your comment had me roll up my sleeves. > > I've noticed that every dependency went out of date, and it's hell to get > pyCrossfade working. So I've created a Dockerfile to nail all these values > down. > > And scripted usage didn't feel solid, so I've created a CLI for easier > interaction. > > Please try the new docker container CLI and with the pycrossfade > crossfade --verbose option, and tell me what you think and what you'd > need? > > (I've only implemented the --verbose flag for crossfade and not > crossfade-many, I'll be updating it based on your feedback) > > — > Reply to this email directly, view it on GitHub > <#8 (comment)>, > or unsubscribe > <https://github.com/notifications/unsubscribe-auth/AACCIQV4DKR7KBP7VACZE4T2EXQQJAVCNFSM6AAAAABSUIYSBKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDKMRZGA3TQNRTGI> > . > You are receiving this because you were mentioned.Message ID: > ***@***.***> >