Removed some unnecessary subtitle screen formatting, increased ASS re…

…solution to match video resolution, updated dependencies to latest whisper etc.

.github/removetriton.patch

@@ -1,41 +1,30 @@
-712d711
-< triton = "2.0.0"
-1206,1232d1204
-< 
-< [[package]]
+1040d1039
+< triton = ">=2.0.0,<3"
+1657d1655
+< triton = {version = "2.1.0", markers = "platform_system == \"Linux\" and platform_machine == \"x86_64\""}
+1685,1709d1682
 < name = "triton"
-< version = "2.0.0"
+< version = "2.1.0"
 < description = "A language and compiler for custom Deep Learning operations"
 < optional = false
 < python-versions = "*"
 < files = [
-<     {file = "triton-2.0.0-1-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:38806ee9663f4b0f7cd64790e96c579374089e58f49aac4a6608121aa55e2505"},
-<     {file = "triton-2.0.0-1-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:226941c7b8595219ddef59a1fdb821e8c744289a132415ddd584facedeb475b1"},
-<     {file = "triton-2.0.0-1-cp36-cp36m-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:4c9fc8c89874bc48eb7e7b2107a9b8d2c0bf139778637be5bfccb09191685cfd"},
-<     {file = "triton-2.0.0-1-cp37-cp37m-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:d2684b6a60b9f174f447f36f933e9a45f31db96cb723723ecd2dcfd1c57b778b"},
-<     {file = "triton-2.0.0-1-cp38-cp38-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:9d4978298b74fcf59a75fe71e535c092b023088933b2f1df933ec32615e4beef"},
-<     {file = "triton-2.0.0-1-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:74f118c12b437fb2ca25e1a04759173b517582fcf4c7be11913316c764213656"},
-<     {file = "triton-2.0.0-1-pp37-pypy37_pp73-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:9618815a8da1d9157514f08f855d9e9ff92e329cd81c0305003eb9ec25cc5add"},
-<     {file = "triton-2.0.0-1-pp38-pypy38_pp73-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:1aca3303629cd3136375b82cb9921727f804e47ebee27b2677fef23005c3851a"},
-<     {file = "triton-2.0.0-1-pp39-pypy39_pp73-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:e3e13aa8b527c9b642e3a9defcc0fbd8ffbe1c80d8ac8c15a01692478dc64d8a"},
-<     {file = "triton-2.0.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:8f05a7e64e4ca0565535e3d5d3405d7e49f9d308505bb7773d21fb26a4c008c2"},
-<     {file = "triton-2.0.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:bb4b99ca3c6844066e516658541d876c28a5f6e3a852286bbc97ad57134827fd"},
-<     {file = "triton-2.0.0-cp36-cp36m-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:47b4d70dc92fb40af553b4460492c31dc7d3a114a979ffb7a5cdedb7eb546c08"},
-<     {file = "triton-2.0.0-cp37-cp37m-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:fedce6a381901b1547e0e7e1f2546e4f65dca6d91e2d8a7305a2d1f5551895be"},
-<     {file = "triton-2.0.0-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:75834f27926eab6c7f00ce73aaf1ab5bfb9bec6eb57ab7c0bfc0a23fac803b4c"},
-<     {file = "triton-2.0.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0117722f8c2b579cd429e0bee80f7731ae05f63fe8e9414acd9a679885fcbf42"},
-<     {file = "triton-2.0.0-pp37-pypy37_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:bcd9be5d0c2e45d2b7e6ddc6da20112b6862d69741576f9c3dbaf941d745ecae"},
-<     {file = "triton-2.0.0-pp38-pypy38_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:42a0d2c3fc2eab4ba71384f2e785fbfd47aa41ae05fa58bf12cb31dcbd0aeceb"},
-<     {file = "triton-2.0.0-pp39-pypy39_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:52c47b72c72693198163ece9d90a721299e4fb3b8e24fd13141e384ad952724f"},
+<     {file = "triton-2.1.0-0-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:66439923a30d5d48399b08a9eae10370f6c261a5ec864a64983bae63152d39d7"},
+<     {file = "triton-2.1.0-0-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:919b06453f0033ea52c13eaf7833de0e57db3178d23d4e04f9fc71c4f2c32bf8"},
+<     {file = "triton-2.1.0-0-cp37-cp37m-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:ae4bb8a91de790e1866405211c4d618379781188f40d5c4c399766914e84cd94"},
+<     {file = "triton-2.1.0-0-cp38-cp38-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:39f6fb6bdccb3e98f3152e3fbea724f1aeae7d749412bbb1fa9c441d474eba26"},
+<     {file = "triton-2.1.0-0-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:21544e522c02005a626c8ad63d39bdff2f31d41069592919ef281e964ed26446"},
+<     {file = "triton-2.1.0-0-pp37-pypy37_pp73-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:143582ca31dd89cd982bd3bf53666bab1c7527d41e185f9e3d8a3051ce1b663b"},
+<     {file = "triton-2.1.0-0-pp38-pypy38_pp73-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:82fc5aeeedf6e36be4e4530cbdcba81a09d65c18e02f52dc298696d45721f3bd"},
+<     {file = "triton-2.1.0-0-pp39-pypy39_pp73-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:81a96d110a738ff63339fc892ded095b31bd0d205e3aace262af8400d40b6fa8"},
 < ]
-1234,1243d1205
+< 
 < [package.dependencies]
-< cmake = "*"
 < filelock = "*"
-< lit = "*"
-< torch = "*"
 < 
 < [package.extras]
+< build = ["cmake (>=3.18)", "lit"]
 < tests = ["autopep8", "flake8", "isort", "numpy", "pytest", "scipy (>=1.7.1)"]
 < tutorials = ["matplotlib", "pandas", "tabulate"]
 < 
+< [[package]]

lyrics_transcriber/transcriber.py

@@ -76,17 +76,21 @@ def __init__(
 
         match video_resolution:
             case "4k":
-                self.video_resolution_num = ("3840", "2160")
+                self.video_resolution_num = (3840, 2160)
                 self.font_size = 250
+                self.line_height = 250
             case "1080p":
-                self.video_resolution_num = ("1920", "1080")
-                self.font_size = 140
+                self.video_resolution_num = (1920, 1080)
+                self.font_size = 120
+                self.line_height = 120
             case "720p":
-                self.video_resolution_num = ("1280", "720")
+                self.video_resolution_num = (1280, 720)
                 self.font_size = 100
+                self.line_height = 100
             case "360p":
-                self.video_resolution_num = ("640", "360")
+                self.video_resolution_num = (640, 360)
                 self.font_size = 50
+                self.line_height = 50
             case _:
                 raise ValueError("Invalid video_resolution value. Must be one of: 4k, 1080p, 720p, 360p")
 
@@ -577,6 +581,8 @@ def create_screens(self):
             if screen is None:
                 self.logger.debug(f"screen is none, creating new LyricsScreen")
                 screen = subtitles.LyricsScreen()
+                screen.video_size = self.video_resolution_num
+                screen.line_height = self.line_height
             if line is None:
                 self.logger.debug(f"line is none, creating new LyricsLine")
                 line = subtitles.LyricsLine()

lyrics_transcriber/utils/subtitles.py

@@ -1,6 +1,6 @@
 from dataclasses import dataclass, field
 from datetime import timedelta
-from typing import Dict, List, Optional
+from typing import Dict, List, Optional, Tuple
 import json
 import itertools
 from pathlib import Path
@@ -13,14 +13,6 @@
 Functions for generating ASS subtitles from lyric data
 """
 
-VIDEO_SIZE = (400, 320)
-LINE_HEIGHT = 30
-
-
-class LyricMarker(IntEnum):
-    SEGMENT_START = 1
-    SEGMENT_END = 2
-
 
 class LyricSegmentIterator:
     def __init__(self, lyrics_segments: List[str]):
@@ -143,16 +135,17 @@ def from_dict(cls, data: dict) -> "LyricsLine":
 class LyricsScreen:
     lines: List[LyricsLine] = field(default_factory=list)
     start_ts: Optional[timedelta] = None
+    video_size: Tuple[int, int] = None
+    line_height: int = None
 
     @property
     def end_ts(self) -> timedelta:
         return self.lines[-1].end_ts
 
     def get_line_y(self, line_num: int) -> int:
-        _, h = VIDEO_SIZE
+        _, h = self.video_size
         line_count = len(self.lines)
-        line_height = LINE_HEIGHT
-        return (h / 2) - (line_count * line_height / 2) + (line_num * line_height)
+        return (h / 2) - (line_count * self.line_height / 2) + (line_num * self.line_height)
 
     def as_ass_events(self, style: ass.ASS.Style) -> List[ass.ASS.Event]:
         return [line.as_ass_event(self.start_ts, self.end_ts, style, self.get_line_y(i)) for i, line in enumerate(self.lines)]
@@ -185,46 +178,6 @@ def default(self, o):
         return super().default(o)
 
 
-def create_screens(logger, lyrics_segments, events_tuples):
-    segments = iter(LyricSegmentIterator(lyrics_segments=lyrics_segments))
-    events = iter(events_tuples)
-    screens: List[LyricsScreen] = []
-    prev_segment: Optional[LyricSegment] = None
-    line: Optional[LyricsLine] = None
-    screen: Optional[LyricsScreen] = None
-
-    try:
-        for event in events:
-            ts = event[0]
-            marker = event[1]
-            if marker == LyricMarker.SEGMENT_START:
-                segment_text: str = next(segments)
-                segment = LyricSegment(segment_text, ts)
-                if screen is None:
-                    screen = LyricsScreen()
-                if line is None:
-                    line = LyricsLine()
-                line.segments.append(segment)
-                if segment_text.endswith("\n"):
-                    screen.lines.append(line)
-                    line = None
-                if segment_text.endswith("\n\n"):
-                    screens.append(screen)
-                    screen = None
-                prev_segment = segment
-            elif marker == LyricMarker.SEGMENT_END:
-                if prev_segment is not None:
-                    prev_segment.end_ts = ts
-        if line is not None:
-            screen.lines.append(line)  # type: ignore[union-attr]
-        if screen is not None and len(screen.lines) > 0:
-            screens.append(screen)  # type: ignore[arg-type]
-    except StopIteration as si:
-        logger.error(f"Reached end of segments before end of events. Events: {list(events)}, lyrics: {list(segments)}")
-
-    return screens
-
-
 def set_segment_end_times(screens: List[LyricsScreen], song_duration_seconds: int) -> List[LyricsScreen]:
     """
     Infer end times of lines for screens where they are not already set.
@@ -254,10 +207,14 @@ def set_screen_start_times(screens: List[LyricsScreen]) -> List[LyricsScreen]:
     return screens
 
 
-def create_styled_subtitles(lyric_screens: List[LyricsScreen], resolution, fontsize) -> ass.ASS:
+def create_styled_subtitles(
+    lyric_screens: List[LyricsScreen],
+    resolution,
+    fontsize,
+) -> ass.ASS:
     a = ass.ASS()
     a.set_resolution(resolution)
-    
+
     a.styles_format = [
         "Name",  # The name of the Style. Case sensitive. Cannot include commas.
         "Fontname",  # The fontname as used by Windows. Case-sensitive.
@@ -278,9 +235,9 @@ def create_styled_subtitles(lyric_screens: List[LyricsScreen], resolution, fonts
         "Outline",  # If BorderStyle is 1,  then this specifies the width of the outline around the text, in pixels. Values may be 0, 1, 2, 3 or 4.
         "Shadow",  # If BorderStyle is 1,  then this specifies the depth of the drop shadow behind the text, in pixels. Values may be 0, 1, 2, 3 or 4. Drop shadow is always used in addition to an outline - SSA will force an outline of 1 pixel if no outline width is given.
         "Alignment",  # This sets how text is "justified" within the Left/Right onscreen margins, and also the vertical placing. Values may be 1=Left, 2=Centered, 3=Right. Add 4 to the value for a "Toptitle". Add 8 to the value for a "Midtitle". eg. 5 = left-justified toptitle
-        "MarginL",  #
-        "MarginR",  #
-        "MarginV",  #
+        "MarginL",  # This defines the Left Margin in pixels. It is the distance from the left-hand edge of the screen.The three onscreen margins (MarginL, MarginR, MarginV) define areas in which the subtitle text will be displayed.
+        "MarginR",  # This defines the Right Margin in pixels. It is the distance from the right-hand edge of the screen.
+        "MarginV",  # MarginV. This defines the vertical Left Margin in pixels. For a subtitle, it is the distance from the bottom of the screen. For a toptitle, it is the distance from the top of the screen. For a midtitle, the value is ignored - the text will be vertically centred
         "Encoding",  #
     ]