BPMtoFPS

This package provides tools to convert music time into video time.

Do you have the timecode for the moment in your song where you want a critical action to take place in the video? Are you looking to create a visualization of your favorite piece of music? Do you want to edit your own music video but need to put down markers on your timeline for where all the important beats are? BPMtoFPS was constructed with these exact needs in mind.

The math involved in this process is not complicated. If you know the exact parameters already you can find countless guides and websites for one-off conversions. For a more programmatic approach, however, these tools can save you time by taking large datasets of time values and converting them to the formats you need.

Inputs

What information about your music do you have on hand? This package accepts the following with examples:

1. Beats (integer) - You know your song is 128 beats per minute, and you need the timing for every 24 beats in your 29.97fps video project.
2. Ticks (integer) - You know your MIDI song file has a 480 tick-per-quarter-note resolution, is an instrument from a slow jam at 86 BPM, and you need to know when the note at the 34,812th tick appears in your 60fps video timeline.
3. Measures (integer) - There's no better transition in all music than the one that occurs between Section V and Section VI in Steve Reich's "Music for 18 Musicians." It occurs at measure 392 of the score (not counting repetitions per bar which makes this marker functionally useless but just go with me on this one). What specific frame in a 60fps video does this transition occur?
4. Timecode (string) - Your song isn't of digital origin or has a shifting BPM, but you pause the music at 37.333 and realize you need a video effect to take place at that exact time in your 23.96fps video and don't feel like doing the math.
5. Video Frames (integer) - The oddball of the group, I threw this in here because, you know what? Sometimes it's good to know when a frame will occur in time. This can be really useful when you're dealing with multiple video effect apps.

Outputs

To assist with constructing your video project, the included functions can deliver results in these formats in any combination, with examples:

1. Frames (integer) - Video compositing software like Adobe After Effects and 3D animation software like Blender cooperate better when you can supply the exact frame number of when things should occur as long as you know the FPS of your project.
2. Timecode (string) - Editing software like Final Cut Pro or Adobe Premiere rely more on timecodes for exact position of video events.
3. Seconds (float) - Plain seconds with as many decimal places as you could possibly need.

Examples of Use

from BPMtoFPS import convert_time

1. convert_time('beats', 'frames', 24, bpm=128, fps=29.97) returns {'frames': 337}
2. convert_time('ticks', 'timecode', 34812, bpm=86, fps=60) returns {'timecode': 50:36}
3. convert_time('measures', 'frames', 392, bpm=208, notes_per_measure=6, fps=60 returns {'frames': 40707}
4. convert_time('timecode', 'timecode', '0:37.333', fps=23.96) returns {'timecode': 37:07}
5. convert_time('ticks', ['frames', 'timecode'], 3840, bpm=192, fps=29.97, ticks_per_beat=360) returns {'frames': 100, 'timecode': '3:10'}
6. convert_time('timecode', ['frames', 'timecode', 'seconds'], '0:45.59', fps=29.97) returns {'frames': 1366, 'timecode': '45:17', 'seconds': 45.59}

Handy Features

Output to Various Formats, Including Seconds!

v1.4.0 now focuses on allowing you to output time in a more Pythonic manner (hello, dictionaries!), but also to allow outputting to plain seconds with whatever number of decimal places Python wants to spit out. Command-line execution is a lot more understandable and fun thanks to capital letters! Running -mVCSpi is a real trip. Full documentation in docs/CLI.md.

Video Frames to Timecode Conversion

As a result of forking myself with v1.3.0, you can now convert frames to timecode... at least until I publish FPStoFPS... maybe.

Custom Fraction

With v1.2.0, you can provide your own threshold for when a resulting decimal in a frame number should be rounded up or down. Because audio has exponentially granular timing in comparison to video (to the tune of 360 or 480 ticks per second with MIDI, or 44,100 or 48,000 samples per second with recorded audio), video editors have to compensate for this by choosing where to place an action on the timeline, either at the frame before or after the exact moment occurs in the audio. By default, BPMtoFPS enforces a 0.75 threshold, so an action occurring at 4:33.67 (4 seconds, 33 frames, and... 67) will be rounded down to 4:33, whereas traditional rounding (0.5) would have rounded up to 4:34. This is purely personal preference and has generally been the threshold I use for my own projects. You can supply your own using the specific functions detailed below (look for the new "fraction" parameter).

Direct Access to Conversion Functions

In addition to custom rounding, if you know what you're doing, you can skip past the mumbo jumbo of the main convert_time() function and instead go directly to the conversion functions themselves. Both input and output conversions are functions you can access directly. See docs/Functions.md for more details.

Command-line Execution

Don't want to pull open a Python console just to retrieve some numbers? BPMtoFPS can be executed from the command line!