Initial commit

demo/README.md

@@ -0,0 +1,65 @@
+# How to Create Videos Like the Demo (52 seconds in just 736kB!)
+
+The demo video showcases Abogen - an all-in-one tool for turning text into something you can see and hear. This guide explains how I created such a small yet effective demonstration video.
+
+## About the Demo
+
+The demo video shows how Abogen:
+- Converts text files (ePub, PDF, text) into audio with synchronized subtitles
+- Uses Kokoro (a powerful text-to-speech engine) to create natural voices
+- Works completely on your computer for privacy and security
+- Offers an easy interface for creating audiobooks and voiceovers
+- Can be used for Instagram, YouTube, TikTok, or any content creation
+
+And it does all this while being only **736kB** for a **52-second video**!
+
+## How I Created This Tiny Video
+
+### What You Need
+
+- A background image (bg.jpg)
+- The subtitle file (.srt) created by Abogen
+- The audio recording (.wav) created by Abogen
+- FFmpeg installed on your computer:
+
+```bash
+# Windows
+winget install ffmpeg
+# MacOS
+brew install ffmpeg
+# Linux
+sudo apt install ffmpeg
+```
+
+### Step 1: Process the Subtitle File
+
+Run this command to process Abogen's subtitle file:
+
+```
+python convert.py abogen_subtitles.srt
+```
+
+This creates a properly formatted subtitle file called "abogen_subtitles_demo.ass" with centered text and appropriate styling.
+
+### Step 2: Create the Video
+
+Run this FFmpeg command to create the tiny video:
+
+```
+ffmpeg -loop 1 -framerate 24 -i bg.jpg -i audio.wav -vf "ass=abogen_subtitles_demo.ass" -c:v libvpx-vp9 -b:v 0 -crf 30 -c:a libopus -shortest demo.webm
+```
+
+That's it! The magic happens because:
+- We use a single static background image instead of many frames
+- The subtitles are stored as text (vector data), not as pixels
+- VP9 video codec with Opus audio provides excellent compression
+
+## For Higher Quality (But Larger) Video
+
+If you need better quality for distribution, use this command instead:
+
+```
+ffmpeg -loop 1 -framerate 24 -i bg.jpg -i audio.wav -vf "ass=abogen_subtitles_demo.ass" -c:v libx264 -preset slow -crf 18 -movflags +faststart -c:a copy -shortest demo.mp4
+```
+
+This creates an MP4 file that's compatible with more devices but larger in size.

demo/abogen.txt

@@ -0,0 +1 @@
+Abogen is your all-in-one tool for turning text into something you can see and hear. Easily convert ePub, PDF, or text file into audio with subtitles that light up every word as it’s spoken. The voice you hear, and the subtitles you see are both created automatically by Abogen using Kokoro, an open-weight TTS model with 82 million parameters. Everything works right on your computer, so your files stay private and secure. The easy interface lets anyone create audiobooks, voiceovers for Instagram, YouTube or TikTok, or just high-quality text-to-speech in seconds. Whether you want to listen, create content, or bring your words to life, Abogen makes it fast, fun, and effortless. Experience your stories, scripts, and ideas in a whole new way with Abogen.

demo/convert.py

@@ -0,0 +1,205 @@
+# FFmpeg commands for creating videos:
+#
+# For WebM (smaller filesize):
+# ffmpeg -loop 1 -framerate 24 -i bg.jpg -i audio.wav -vf "ass=your_subtitle_demo.ass" -c:v libvpx-vp9 -b:v 0 -crf 30 -c:a libopus -shortest demo.webm
+# 
+# For MP4 (higher quality):
+# ffmpeg -loop 1 -framerate 24 -i bg.jpg -i audio.wav -vf "ass=your_subtitle_demo.ass" -c:v libx264 -preset slow -crf 18 -movflags +faststart -c:a copy -shortest demo.mp4
+#
+# Note: Replace 'audio.wav' with your audio file and 'your_subtitle_demo.ass' with your processed subtitle file
+
+
+import re
+import sys
+from datetime import timedelta
+
+def parse_time(s):
+    # For ASS format
+    if ':' in s and '.' in s:
+        h, m, s = s.split(':')
+        sec, cs = s.split('.')
+        return timedelta(hours=int(h), minutes=int(m), seconds=int(sec), milliseconds=int(cs) * 10)
+    # For SRT format (00:00:00,000)
+    elif ':' in s and ',' in s:
+        parts = s.split(':')
+        h = int(parts[0])
+        m = int(parts[1])
+        sec_parts = parts[2].split(',')
+        sec = int(sec_parts[0])
+        ms = int(sec_parts[1])
+        return timedelta(hours=h, minutes=m, seconds=sec, milliseconds=ms)
+    return None
+
+def format_time(t):
+    total_seconds = int(t.total_seconds())
+    cs = int((t.total_seconds() - total_seconds) * 100)
+    h = total_seconds // 3600
+    m = (total_seconds % 3600) // 60
+    s = total_seconds % 60
+    return f"{h}:{m:02}:{s:02}.{cs:02}"
+
+# Desired script info and style
+DESIRED_SCRIPT_INFO = """[Script Info]
+Title: Centered Subs
+ScriptType: v4.00+
+PlayResX: 1280
+PlayResY: 720
+"""
+
+DESIRED_STYLE = """[V4+ Styles]
+Format: Name, Fontname, Fontsize, PrimaryColour, SecondaryColour, OutlineColour, BackColour, Bold, Italic, Underline, StrikeOut, ScaleX, ScaleY, Spacing, Angle, BorderStyle, Outline, Shadow, Alignment, MarginL, MarginR, MarginV, Encoding
+Style: Default,Arial,60,&H00FFFFFF,&H000000FF,&H00000000,&H64000000,-1,0,0,0,100,100,0,0,1,2,0,5,10,10,10,1
+"""
+
+def process_subtitle_file(input_file):
+    output_file = input_file.replace('.ass', '_demo.ass').replace('.srt', '_demo.ass')
+    if output_file == input_file:
+        output_file = f"{input_file}_demo.ass"
+
+    try:
+        with open(input_file, "r", encoding="utf-8") as f:
+            lines = f.readlines()
+    except FileNotFoundError:
+        print(f"❌ Error: Input file '{input_file}' not found")
+        sys.exit(1)
+    except Exception as e:
+        print(f"❌ Error reading file: {e}")
+        sys.exit(1)
+
+    # Check if it's an SRT file
+    is_srt = input_file.lower().endswith('.srt')
+    
+    if is_srt:
+        return convert_srt_to_ass(input_file, output_file, lines)
+    
+    # Process ASS file
+    header = []
+    events = []
+    in_events = False
+    script_info_found = False
+    styles_found = False
+
+    for line in lines:
+        if line.strip().startswith("[Script Info]"):
+            script_info_found = True
+            header.append(DESIRED_SCRIPT_INFO)
+            continue
+        elif line.strip().startswith("[V4+ Styles]"):
+            styles_found = True
+            header.append(DESIRED_STYLE)
+            continue
+        elif line.strip().startswith("[Events]"):
+            in_events = True
+            header.append(line)
+        elif not in_events:
+            if not script_info_found and not styles_found:
+                header.append(line)
+        else:
+            if line.startswith("Dialogue:"):
+                events.append(line)
+            else:
+                header.append(line)
+
+    parsed_events = []
+    for line in events:
+        match = re.match(r"Dialogue: \d,([^,]+),([^,]+),", line)
+        if not match:
+            parsed_events.append((None, None, line))
+            continue
+        start = parse_time(match.group(1))
+        end = parse_time(match.group(2))
+        parsed_events.append((start, end, line))
+
+    # Cut overlaps
+    fixed_lines = []
+    for i in range(len(parsed_events)):
+        start, end, line = parsed_events[i]
+        if start is None:
+            fixed_lines.append(line)
+            continue
+
+        # Check for next subtitle
+        if i + 1 < len(parsed_events):
+            next_start, _, _ = parsed_events[i + 1]
+            if end > next_start:
+                end = next_start  # cut current subtitle to stop at next one's start
+
+        fixed_line = re.sub(r"Dialogue: \d,[^,]+,[^,]+,", 
+                          f"Dialogue: 0,{format_time(start)},{format_time(end)},", 
+                          line, count=1)
+        fixed_lines.append(fixed_line)
+
+    try:
+        with open(output_file, "w", encoding="utf-8") as f:
+            f.writelines(header)
+            f.writelines(fixed_lines)
+        print(f"✅ Successfully processed. Output file: {output_file}")
+    except Exception as e:
+        print(f"❌ Error writing output file: {e}")
+        sys.exit(1)
+
+def convert_srt_to_ass(input_file, output_file, lines):
+    """Convert SRT format to ASS format."""
+    events = []
+    subtitle_blocks = []
+    current_block = []
+    
+    # Parse SRT file
+    for line in lines:
+        line = line.strip()
+        if not line:
+            if current_block:
+                subtitle_blocks.append(current_block)
+                current_block = []
+        else:
+            current_block.append(line)
+    
+    # Don't forget the last block
+    if current_block:
+        subtitle_blocks.append(current_block)
+    
+    # Process subtitle blocks
+    for block in subtitle_blocks:
+        if len(block) < 3:
+            continue  # Invalid block
+            
+        # Skip the subtitle number
+        timing_line = block[1]
+        
+        # Parse timing information
+        timing_match = re.match(r'(\d+:\d+:\d+,\d+)\s+-->\s+(\d+:\d+:\d+,\d+)', timing_line)
+        if not timing_match:
+            continue
+            
+        start_time = parse_time(timing_match.group(1))
+        end_time = parse_time(timing_match.group(2))
+        
+        # Combine text lines
+        text = "\\N".join(block[2:])
+        
+        # Create ASS Dialogue line
+        dialogue_line = f"Dialogue: 0,{format_time(start_time)},{format_time(end_time)},Default,,0,0,0,,{text}\n"
+        events.append(dialogue_line)
+    
+    # Create ASS file
+    try:
+        with open(output_file, "w", encoding="utf-8") as f:
+            f.write(DESIRED_SCRIPT_INFO)
+            f.write(DESIRED_STYLE)
+            f.write("[Events]\n")
+            f.write("Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text\n")
+            f.writelines(events)
+        print(f"✅ Successfully converted SRT to ASS. Output file: {output_file}")
+        return True
+    except Exception as e:
+        print(f"❌ Error writing output file: {e}")
+        sys.exit(1)
+
+if __name__ == "__main__":
+    if len(sys.argv) < 2:
+        print("❌ Error: No input file specified")
+        print(f"Usage: {sys.argv[0]} <input_file.ass|input_file.srt>")
+        sys.exit(1)
+    
+    input_file = sys.argv[1]
+    process_subtitle_file(input_file)

demo/demo.srt

@@ -0,0 +1,480 @@
+1
+00:00:00,350 --> 00:00:00,962
+Abogen
+
+2
+00:00:00,962 --> 00:00:01,100
+is
+
+3
+00:00:01,100 --> 00:00:01,300
+your
+
+4
+00:00:01,300 --> 00:00:01,887
+all-in-one
+
+5
+00:00:01,887 --> 00:00:02,250
+tool
+
+6
+00:00:02,250 --> 00:00:02,375
+for
+
+7
+00:00:02,375 --> 00:00:02,762
+turning
+
+8
+00:00:02,762 --> 00:00:03,162
+text
+
+9
+00:00:03,162 --> 00:00:03,375
+into
+
+10
+00:00:03,375 --> 00:00:03,725
+something
+
+11
+00:00:03,725 --> 00:00:03,825
+you
+
+12
+00:00:03,825 --> 00:00:04,025
+can
+
+13
+00:00:04,025 --> 00:00:04,312
+see
+
+14
+00:00:04,312 --> 00:00:04,450
+and
+
+15
+00:00:04,450 --> 00:00:05,450
+hear.
+
+16
+00:00:05,450 --> 00:00:05,900
+Easily
+
+17
+00:00:05,900 --> 00:00:06,337
+convert
+
+18
+00:00:06,337 --> 00:00:07,187
+ePub,
+
+19
+00:00:07,187 --> 00:00:08,250
+PDF,
+
+20
+00:00:08,250 --> 00:00:08,412
+or
+
+21
+00:00:08,412 --> 00:00:08,800
+text
+
+22
+00:00:08,800 --> 00:00:09,162
+file
+
+23
+00:00:09,162 --> 00:00:09,425
+into
+
+24
+00:00:09,425 --> 00:00:10,037
+audio
+
+25
+00:00:10,037 --> 00:00:10,237
+with
+
+26
+00:00:10,237 --> 00:00:10,875
+subtitles
+
+27
+00:00:10,875 --> 00:00:11,087
+that
+
+28
+00:00:11,087 --> 00:00:11,287
+light
+
+29
+00:00:11,287 --> 00:00:11,487
+up
+
+30
+00:00:11,487 --> 00:00:11,824
+every
+
+31
+00:00:11,824 --> 00:00:12,175
+word
+
+32
+00:00:12,175 --> 00:00:12,300
+as
+
+33
+00:00:12,300 --> 00:00:12,449
+it’s
+
+34
+00:00:12,449 --> 00:00:13,787
+spoken.
+
+35
+00:00:13,787 --> 00:00:13,925
+The
+
+36
+00:00:13,925 --> 00:00:14,275
+voice
+
+37
+00:00:14,275 --> 00:00:14,425
+you
+
+38
+00:00:14,425 --> 00:00:15,212
+hear,
+
+39
+00:00:15,212 --> 00:00:15,337
+and
+
+40
+00:00:15,337 --> 00:00:15,437
+the
+
+41
+00:00:15,437 --> 00:00:16,024
+subtitles
+
+42
+00:00:16,024 --> 00:00:16,162
+you
+
+43
+00:00:16,162 --> 00:00:16,487
+see
+
+44
+00:00:16,487 --> 00:00:16,649
+are
+
+45
+00:00:16,649 --> 00:00:16,975
+both
+
+46
+00:00:16,975 --> 00:00:17,449
+created
+
+47
+00:00:17,449 --> 00:00:18,362
+automatically
+
+48
+00:00:18,362 --> 00:00:18,550
+by
+
+49
+00:00:18,550 --> 00:00:19,074
+Abogen
+
+50
+00:00:19,074 --> 00:00:19,425
+using
+
+51
+00:00:19,425 --> 00:00:20,512
+Kokoro,
+
+52
+00:00:20,512 --> 00:00:20,637
+a
+
+53
+00:00:20,637 --> 00:00:21,237
+powerful
+
+54
+00:00:21,237 --> 00:00:22,050
+text-to-speech
+
+55
+00:00:22,050 --> 00:00:23,225
+engine.
+
+56
+00:00:23,225 --> 00:00:23,750
+Everything
+
+57
+00:00:23,750 --> 00:00:24,062
+works
+
+58
+00:00:24,062 --> 00:00:24,312
+right
+
+59
+00:00:24,312 --> 00:00:24,412
+on
+
+60
+00:00:24,412 --> 00:00:24,524
+your
+
+61
+00:00:24,524 --> 00:00:25,437
+computer,
+
+62
+00:00:25,437 --> 00:00:25,625
+so
+
+63
+00:00:25,625 --> 00:00:25,812
+your
+
+64
+00:00:25,812 --> 00:00:26,225
+files
+
+65
+00:00:26,225 --> 00:00:26,524
+stay
+
+66
+00:00:26,524 --> 00:00:27,112
+private
+
+67
+00:00:27,112 --> 00:00:27,274
+and
+
+68
+00:00:27,274 --> 00:00:28,337
+secure.
+
+69
+00:00:28,725 --> 00:00:28,862
+The
+
+70
+00:00:28,862 --> 00:00:29,225
+easy
+
+71
+00:00:29,225 --> 00:00:29,824
+interface
+
+72
+00:00:29,824 --> 00:00:30,112
+lets
+
+73
+00:00:30,112 --> 00:00:30,637
+anyone
+
+74
+00:00:30,637 --> 00:00:31,062
+create
+
+75
+00:00:31,062 --> 00:00:32,299
+audiobooks,
+
+76
+00:00:32,299 --> 00:00:33,075
+voiceovers
+
+77
+00:00:33,075 --> 00:00:33,237
+for
+
+78
+00:00:33,237 --> 00:00:34,325
+Instagram,
+
+79
+00:00:34,325 --> 00:00:34,962
+YouTube
+
+80
+00:00:34,962 --> 00:00:35,137
+or
+
+81
+00:00:35,137 --> 00:00:35,950
+TikTok,
+
+82
+00:00:35,950 --> 00:00:36,100
+or
+
+83
+00:00:36,100 --> 00:00:36,325
+just
+
+84
+00:00:36,325 --> 00:00:37,024
+high-quality
+
+85
+00:00:37,024 --> 00:00:37,850
+text-to-speech
+
+86
+00:00:37,850 --> 00:00:38,000
+in
+
+87
+00:00:38,000 --> 00:00:39,187
+seconds.
+
+88
+00:00:39,187 --> 00:00:39,475
+Whether
+
+89
+00:00:39,475 --> 00:00:39,625
+you
+
+90
+00:00:39,625 --> 00:00:39,825
+want
+
+91
+00:00:39,825 --> 00:00:39,924
+to
+
+92
+00:00:39,924 --> 00:00:40,662
+listen,
+
+93
+00:00:40,662 --> 00:00:41,000
+create
+
+94
+00:00:41,000 --> 00:00:41,774
+content,
+
+95
+00:00:41,774 --> 00:00:41,924
+or
+
+96
+00:00:41,924 --> 00:00:42,137
+bring
+
+97
+00:00:42,137 --> 00:00:42,250
+your
+
+98
+00:00:42,250 --> 00:00:42,575
+words
+
+99
+00:00:42,575 --> 00:00:42,700
+to
+
+100
+00:00:42,700 --> 00:00:43,512
+life,
+
+101
+00:00:43,512 --> 00:00:44,075
+Abogen
+
+102
+00:00:44,075 --> 00:00:44,350
+makes
+
+103
+00:00:44,350 --> 00:00:44,525
+it
+
+104
+00:00:44,525 --> 00:00:45,299
+fast,
+
+105
+00:00:45,299 --> 00:00:45,825
+fun,
+
+106
+00:00:45,825 --> 00:00:45,950
+and
+
+107
+00:00:45,950 --> 00:00:47,275
+effortless.
+
+108
+00:00:47,275 --> 00:00:47,962
+Experience
+
+109
+00:00:47,962 --> 00:00:48,075
+your
+
+110
+00:00:48,075 --> 00:00:48,812
+stories,
+
+111
+00:00:48,812 --> 00:00:49,350
+scripts,
+
+112
+00:00:49,350 --> 00:00:49,500
+and
+
+113
+00:00:49,500 --> 00:00:50,075
+ideas
+
+114
+00:00:50,075 --> 00:00:50,174
+in
+
+115
+00:00:50,174 --> 00:00:50,299
+a
+
+116
+00:00:50,299 --> 00:00:50,662
+whole
+
+117
+00:00:50,662 --> 00:00:50,850
+new
+
+118
+00:00:50,850 --> 00:00:51,150
+way
+
+119
+00:00:51,150 --> 00:00:51,299
+with
+
+120
+00:00:51,299 --> 00:00:52,424
+Abogen.
+