Aria The Animation (S1) BD Subtitle Indonesia Batch, Download Aria The Animation (S1) BD Subtitle Indonesia Batch Lengkap, Aria The Animation (S1) BD Indo. 1st series ARIA The ANIMATION 2nd series ARIA The NATURAL 3rd series ARIA The ORIGINATION. Movie ARIA The CREPUSCOLO 「ARIA The CREPUSCOLO」Blu-ray. A conclusive study of Aria The Animation was not performed, but some changes spotted include correcting hair color mistakes, fixing flat-looking faces, and altering clothing. Lyric deviation The lyrics for Just For You misses a few phrases that were sung.

1. Aria The Animation Dvd
2. Bd Aria Ii
3. Aria The Animation Blu Ray
4. Watch Aria The Animation Dub
5. Aria The Animation Dub

Aria The Animation BD Batch Subtitle Indonesia

Alur Cerita:

Melayang dengan tenang ke Neo Venezia, sebuah kota di planet Aqua (sebelumnya dikenal sebagai Mars). Pada abad ke-24, manusia telah menemukan cara untuk menjajah planet yang sebelumnya tidak dapat dihuni. Meski terdengar futuristik, Neo Venezia masih penuh dengan keindahan pedesaan; gondola di kanal dan saluran air yang lebar adalah moda transportasi utama. Kota itu sendiri adalah replikasi setia dari Manhome (planet yang sebelumnya dikenal sebagai Bumi) Venesia.

Untuk memastikan bahwa penduduk dan turis sama-sama mendapatkan yang terbaik dari banyak keajaiban Neo Venezia, perusahaan yang menawarkan wisata berpemandu melalui gondola dibentuk, salah satunya bernama Aria Company.

Ini adalah tempat kerja Akari Mizunashi, seorang remaja berjiwa bebas dari Manhome yang sekarang menjadi Undine pemula (gelar yang diberikan kepada pemandu wisata). Bergabunglah dengan Akari saat dia berkenalan secara dekat dengan Undine lainnya, turis, penduduk Neo Venezia, dan bahkan kota itu sendiri, belajar banyak pelajaran hidup yang berharga di sepanjang jalan, seperti kebenaran indah bahwa ada hal-hal seperti keajaiban buatan manusia.

Terima Kasih untuk kalian semua yang masih setia download anime di animebatch.org hingga saat ini berkat kalian semua situs animebatch.org menjadi situs download anime terbaik dan terbesar di indonesia.

Contents

Download

You can download this anime series via BitTorrent by using the “DL” links on this page: http://tracker.weedy.ca:8394/index.html?search=P79068

I encourage you to try the compact edition. =)

Introduction

Thank you for downloading this fansub of Aria The Animation by Project79068. In this document, you will find information about this fansub and how it was made.

Why did Nayuki spend the time to write this up, this thing that can be compared to a long drunken rant? Because we want to set the highest standard for fansubbing and show how seriously it can be taken.

Mission statement

The goals of Project79068 include:

• Empowering users and developers
• Disseminating useful information, techniques, and tools
• Setting a high standard for quality
• Making things easy to understand

Acknowledgements

This project ran from 2006-11-25-Sat to 2007-09-11-Tue. Hundreds of hours of actual hard work were put in, mostly by Nayuki.

Staff

• Project founder
• Obtaining DVD images
• Encoding video and audio
• Ripping subtitles
• Timing subtitles
• Identifying speakers
• Timing karaokes
• Verifying dialogue and song translations
• Editing English text
• Documenting (i.e., this page)
• Quality checking (e.g. for encoding, timing, karaoke timing, speakers’ names, documentation)
• And everything else not mentioned here

I could have relaxed and watched some 500 half-hour episodes in the time it took to fan-subtitle this series. Was it worth my time? Not really. I happened to have some free time, so I did this just once to give something to the world and make a statement about my skills at the same time.

Looking back, I think it’s a bit scary that I possess nearly all the skills needed to produce a high quality fansub. It was essentially a one-person job. But it would’ve been nice to have better skills in translation, more advanced raw grabbing (Share, Winny, IRC, etc.), choosing styles and fonts artistically, and web design.

Well... if I were to do another fansub (zOMG I just contradicted myself!), it might take less time since I’ve got the procedures and infrastructure more or less figured out.

Finally, let me be the first to say this: This fansub is cold, hard proof that I am insane. (I mean that in a good and bad way.)

• Editing English text
• Quality checking

Aria The Animation Dvd

This isn’t my fault! I have a charm of protection against any blame! Chosen only for the supreme levels of pedantry to which I may periodically ascend, and the consummate skill in the English tongue, I was able to nitpick the translation to hell and back.

• Ripping subtitles
• Distribution

Nayuki did all the work. Don’t look at me. I didn’t touch anything. I’m just the little troll who kept bugging Nayuki to hurry up and finish.

Other

Our work is heavily based on the subtitles by Crystal Nova fansubs, which were used without permission or notification. Simply put, without their work, we would have no fansub. (Or perhaps we’d have a raw encode with crappy and incomplete translations.)

Without peer-to-peer file sharing networks and their generous and resourceful users, none of this would have been possible. Getting the DVD ISOs and distributing the finished fansub would both be highly impractical.

• x264 revision 667 (2007-07-18)
• VirtualDubMod 1.5.10.2 (only to preview, not to encode)
• AVISynth 2.5.7
• DGMPGDec 1.4.8
• oggdropXPd 1.8.9 using libvorbis 1.2.0
• mkvmerge 2.1.0
• Java 1.5 (to compile and run custom text processing programs)
• DVD Decrypter 3.5.4.0 (to extract PGCs from VOBs and demultiplex their streams)
• foobar2000 0.8.3 (to decode the audio)
• Notepad 5.1 (to view and edit scripts, source code, etc.)
• grep
• diff

WWWJDIC is a free online Japanese-English dictionary. It’s very thorough, covering words, expressions, verb conjugations, kanji, and more. Nayuki heavily relied on this resource to verify and create translations.

This person single-handedly re-subbed Mahou Shoujo Lyrical Nanoha and A’s (getting DVD ISOs from Share, encoding, manually ripping subtitles from Triad’s TV fansubs of Nanoha, styling), inspiring Nayuki to do the same for Aria The Animation.

Features

The high efficiency, state-of-the-art MPEG-4 Part 10 AVC/H.264 codec is used. Almost every single option that enhances coding efficiency was enabled (at a cost to encoding and possibly decoding speed).

The video was only lightly processed. Black borders were cropped away, then the video was resized to 640×480 using the Lanczos resampler. (Lanczos preserves more detail than the bilinear and bicubic resamplers.) The QP for P frames is 16 for the standard edition episodes and commercials, 20 for the other bonus features, and 24 or 25 for the compact edition episodes. In the spirit of consistency, no attempt was made to increase the quantization for the ending animation.

Xiph.Org Vorbis was chosen. AAC was considered, but the two popular encoders that were available – FAAC and Nero Digital Audio+ – were unsatisfactory. Their frequency responses were not as good as Vorbis’. They added padding and thus did not preserve the exact start offset and length. Vorbis does not suffer from these problems. Furthermore, it is free software (FOSS), offsetting any small potential performance gains from AAC with greater accessibility across platforms and across time.

In the standard edition, q 8 was used; in the compact edition, q 3 was used. For the compact edition, P79068sincresample was used to downsample from 48 kHz to 32 kHz. It is a zero-phase filter with asymptotically perfect frequency domain preservation and anti-aliasing. A comparable off-the-shelf alternative is any program that uses libsamplerate, a.k.a.Secret Rabbit Code.

MPEG-4 AVC and Xiph.Org Vorbis are highly efficient codecs, and the compact edition demonstrates this fact spectacularly: Despite the average size of each episode being 54 MiB, even videophiles and audiophiles will be hard-pressed to find artifacts and degradations.

Note that this efficiency is partly attributed to a clean, very high quality source. If a low quality source were used, a larger part of the encode would be wasted toward preserving artifacts – rather than useful data – in the source.

The soft subtitles are provided in 3 versions:

• ASS (default)
• ASS, without karaoke
• SRT (without notes or lyrics)

In the compact edition, ASS, without karaoke is renamed to ASS and is the default stream. ASS is renamed to ASS, with karaoke.

While almost all fansubbers use hard-subbed karaokes, we don’t know how to generate those fancy animation effects. We used ASS soft-subbed karaokes, which we feel are sufficiently functional.

Note: To display this using VSFilter/DirectVobSub, you may need to disable subpicture buffering.

The subtitles are typeset in Helvetica, a highly robust and legible sans-serif font with more aesthetic shapes than Arial.

Every video clip on the DVDs have been encoded and released. However, with the exception of the commercials, the bonus features have not been translated and subtitled.

All the live-action bonus features were deinterlaced using an adaptive bob-weave filter (LeakKernelBob) to yield full-resolution and full-motion video (720×480 @ 59.94 Hz). However, this comes at a cost of approximately doubling the output file size and playback CPU usage.

For easy navigation, each episode is divided into these chapters: Prologue, Part A, Part B, Ending, Next Navigation. As an example, you can use this to easily skip the ending theme.

Original kanji lyrics (verbatim from CD booklets) is a standard feature in fansubs, intended for those who can read some Japanese.

Our subtitles were originally taken from Crystal Nova’s fansub work. The translations were then verified to the best of our ability, with many instances where a number of minutes were spent re-translating and editing the wording of just a single dialogue line.

By popular convention, name suffixes were left intact because they suggest about relationships and statuses. Akira’s trademark “Suwa!” was also preserved, but it’s too bad dekkai couldn’t be ;-). On the downside, this may confuse and discourage newcomers to fansubbed anime.

Lesser features

The start and end times (before padding) of dialogue events are precise to the millisecond and accurate to a few centiseconds. If timing were done using a waveform view instead of a spectrogram, the accuracy might be several times worse.

American English was chosen over British English because it is an acceptable de facto standard. Some differences include color/colour, realize/realise, and different from/different to.

Proper characters are used, including curved opening/closing quotation marks and apostrophes, accented characters, and en/em dashes. In addition to that, italics are used.

As much as possible, the principle of consistency is applied where it is appropriate. For example: The use of punctuation, interjections, hesitations, etc. is consistent, in the sense that the same mood will yield the same wording. The way lines are split based on the speaking pace is consistent. The vocal features used as timing edges are consistent.

The video is not hard-subtitled for signs, karaokes, or shameless credits. The audio is also clean, of course. (Why would anybody edit the audio anyway?)

A package of the files used to produce this fansub was released publicly. It includes things like:

• Final standalone subtitle files
• Subtitle scripts in lightweight format
• Karaoke scripts in lightweight format
• AVISynth scripts
• x264 encoding logs
• Custom processing and visualization programs written for this project
• Crystal Nova’s subtitle scripts (manually ripped)
• Difference between Crystal Nova’s scripts and our scripts
• Fonts used
• Scanned CD booklet images with song lyrics
• Miscellaneous utilities like the JavaScripter and sincresample

This package open-sources the intermediate files and programs used to produce this fansub. Off-the-shelf programs such as x264 are not included. It is hoped that the package will be useful for those interested in the internals of the fansubbing process, and that future fansubbers can learn from our techniques and experiences – both positive or negative ones.

The package was released only in the bonus features batch of files. File information:

• File name: Project79068 - Aria The Animation - Project files.7z
• Size: 6.0 MiB
• CRC-32 hash: D91FBFC5
• MD5 hash: 1D2037A57791AE787D40FF3F08C876C3
• SHA-1 hash: 338360EC878A9649E23B1417339DB4206EAB7451
• Whirlpool hash: 0F6456CB35DEE8CD6EAE5B15850C394E C24300F5B6CDBF7D7D17C4F28D3EB220 CA5C5963BDDE81D7B37A7438922F286C 6A1B2A4DF33B5327050350898A87B6FA

The starting and ending edges of the video clips are trimmed as much as possible to remove black frames and silence. Specifically: If the edge is a scene cut, then all black frames are discarded. Otherwise if the edge is a fade, then one black frame is retained. In either case, the trimming must not result in discarding meaningful audio.

In each episode and commercial, these fonts are embedded: Neue Helvetica Bold, Neue Helvetica Bold Italic, MS Gothic. Episode 09 also includes Arial Bold. In the compact edition, MS Gothic is not embedded because of its hefty cost of 8 MiB.

All the embedded Vorbis audio streams have Replay Gain information (track/radio). For the episodes, the Replay Gain hovers around +5 dB.

Techniques

The Cool Edit Pro 2.0 spectral display was used with these settings:

• Frequency resolution: 1024 bands (for 24 kHz of bandwidth)
• Windowing function: Blackmann-Harris
• Window width: 100%
• Plot style: Logarithmic energy plot, range of 150 dB
• Horizontal (time) resolution: ~8.2 ms per pixel (usually displaying exactly 10 s in 1224 pixels)
• Vertical (frequency) resolution: ~40 Hz per pixel (usually displaying about 0–14 kHz in exactly 379 pixels)

Nayuki is fully aware of the inherent time-frequency uncertainty trade-off in Fourier transforms.

Each dialogue event’s original end time is the exact moment when the voice drops off. This is then padded by the algorithm in the subpad program.

When syllables in songs are timed, their start and end times are measured as usual. Then, these timing edges are snapped snapped to beats or simple fractions thereof, with the beat length determined by the tempo.

Note that even short rests are properly timed, accurately indicating when the voice ends.

By playing a song at a slower tempo, karaoke timing errors are more apparent. Note that during verification, snap style karaoke is used instead of fill style.

For easy editing (less boilerplate, lighter syntax) in plain text editors, new file formats were created during this project. Programs to process data in these formats were also created, of course. To illustrate, compare how a dialogue event is expressed in ASS and in the custom format:

• ASS: Dialogue: 0,0:00:11.36,0:00:12.70,Aika,Aika,0000,0000,0000,Embarrassing remarks aren’t allowed.
• Custom: 11365 <tab> 12704 <tab> Aika <tab> Embarrassing remarks aren’t allowed.

A method of visualizing syllable lengths was created to help find syllable timing errors in songs. It has similarities to piano roll formats.

The tempo of each song was calculated by measuring beats in the spectrogram and then inputting the timings into the beat counter program. The program uses linear regression, which gives more robust results than the simple number of beats ÷ time span method. However, about a week before the release of this fansub, Nayuki found an even more accurate method. (It was too late to use it.)

In episode 07, subtractive cancellation was used to uncover soft dialogue drowned in the loud opening song. In episode 11, an insert song was subtracted. In episode 13, parts of the opening song were subtracted. In all cases, the procedure enhanced the power difference between the speech and the music by a few decibels, making timing easier.

The JavaScripter is an HTML page that allows quickly writing and running JavaScript programs to generate, process, and visualize data.

Processing audio in 16-bit integer is good enough, but 32-bit floating-point is even better: No overflow/saturation (clipping) or underflow (zeroing) under sane processing conditions, less quantization noise (at least as good as 25-bit integer).

Bd Aria Ii

The fact that the compact edition is so close to 700 MiB is no accident. A simple brute force algorithm was used to find the set of files to form the optimal packing. Ahh, the benefits of computer science...

Source quirks

Having worked with the source material for so long while making many intensive passes over it, abnormalities in it were bound to be noticed.

The animated content, including the commercials, is almost entirely progressive but fades are interlaced and a few scenes unexpectedly jump into interlace mode. The live-action bonus features are interlaced.

The video picture for the anime appears somewhat soft. For the sake of not applying unnecessary processing, no attempt was made to sharpen the image.

Although the audio bandwidth is 24 kHz, there is essentially nothing above 20.45 kHz, leaving the top 3.55 kHz unused. However, there are some strange time-domain pulses (which span the entire frequency range) – and they are not caused by clipping. (Maybe they’re watermarks!)

DVD Decrypter indicated that the audio stream of each video clip had an offset of −66 ms, but measurements suggested that no offset was needed. Synchronization was confirmed mostly by sound effects associated with visuals, e.g.: blinking marks around heads, volume changes in background noise on scene changes, and sharp sound effects associated with actions (telephone ringing, objects colliding).

Synchronization between the lip movement and the audio speech is rough. It looks like it’s within ±100 ms or so. But it’s not noticeable in a casual viewing.

A common phenomenon in anime is that some things are redrawn for the DVD release that comes a few months after the first TV airing. A conclusive study of Aria The Animation was not performed, but some changes spotted include correcting hair color mistakes, fixing flat-looking faces, and altering clothing.

The lyrics for Just For You misses a few phrases that were sung. However, it is a common practice to leave out such side remarks from officially published lyrics.

There were some instances where it was not possible to identify the speaker or the dialogue with decent confidence. Also, some lines were hard to time accurately, usually because the spectrogram was usually too cluttered so timing had to be done by ear.

There are some minor spelling, grammar, and word choice mistakes. There are a few translation errors, such as getting the wrong name. But overall, Crystal Nova’s work was an excellent source to base our subtitles on.

All the episode title screens were consistent except for the one in episode 09. Not only did that one come before the opening, it had a slightly different length and fade offset.

In episode 07, three subtle details of the people’s clothing change between frame 17641 and frame 17642 (both 0-based) (~09:48).

Miscellaneous notes

A slightly stripped down version of this about page is embedded in each Matroska file released. File hashes cannot be included, of course, because they would create a circular dependency.

Each dialogue event includes the name of its speaker. This makes it possible to use different styles for each speaker, and to process the subtitles with automated tools such as a lead-out padder.

All the text styles use Neue Helvetica Bold at the same size, except for kanji lyrics (MS Gothic Bold) and translation notes (smaller size). For each character, the text color is derived from their hair color, which was inspired by the anime fansub group Static-Subs.

The official Japanese title of this series is ARIA The ANIMATION. It is decapitalized here since the words are not abbreviations, and CAPITAL LETTERS ARE TIRESOME. The title can also be interpreted as Aria: The Animation.

To keep the punctuation sane, the subtitles lose much of the information from vocal intonation. Please listen to the audio track as you watch the anime.

Ending up with a script as close to Crystal Nova’s as possible was not a goal. When changes needed to be made, they were made. An automatically generated difference analysis between our script and Crystal Nova’s is included in the project files package.

The raw AC-3 audio (192 kb/s, stereo) could have been spliced, offset, and then multiplexed into the finished fansub. Instead, we chose to re-encode to Vorbis (at a tiny loss of quality) to get resulting audio clips that are offset-less and length-exact (to the sample). In addition, Vorbis is a more open format than AC-3.

A frame rate of exactly 30000/1001 Hz was blindly assumed (60000/1000 Hz for deinterlaced bonus features), instead of close alternatives like 30 Hz, 29.97 Hz, or 0.033367 μs per frame. This choice is reflected in the audio lengths, which match the video lengths within 1 audio sample if this frame rate is assumed.

Here is the set of command-line options used in x264 (revision 667, 2007-07-18) to encode all the video clips: --min-keyint 1 --bframes 5 --b-pyramid --ref 20 --partitions all --direct auto --direct-8x8 -1 --weightb --me umh --merange 32 --subme 7 --mixed-refs --bime --b-rdo --8x8dct --trellis 2 --no-fast-pskip --no-dct-decimate

In addition to that are the options that varied for different files:

• Standard edition episodes and commercials: --qp 16 --keyint 100 --sar 8:9 --fps 30000/1001
• Standard edition bonus features: --qp 20 --keyint 200 --sar 8:9 --fps 60000/1001
• Compact edition episodes: --qp 24 --keyint 300 --sar 1:1 --fps 30000/1001 or QP 25

To preserve characters disallowed in file names on the Windows platform, they were escaped using percent encoding (as used in URIs). So %3A means colon (:), and %22 means quotation mark ('). This introduces some problems of its own, though.

Steps not taken

Although these ideas are within reach, they are time-consuming and yield insufficient benefit, so they have not been implemented.

It would be useful to transcribe the raw Japanese dialogue for reference, so that making or verifying the translation does not require listening to the audio. The episodes have not been transcribed, but the two short commercials have been transcribed perfectly. We do not understand every word uttered, though, so only a partial transcription (for the episodes) is possible.

Other than the new enumeration for alignments and a new syntax smooth-fill karaoke (kf), this fansub does not use ASS’ more interesting new features like fading and moving subtitles.

In addition to naming the speaker for each event, it is also possible to indicate its type. For example: speech, thought, narration, flashback.

In the anime, fade-ins, fade-outs, and cross-fades are interlaced. We decided against correcting this problem because there are too many fades and this artifact is sufficiently subtle.

Because we had no automated way of detecting scene cuts, no attempt was made to snap the start or end times of subtitle events to nearby scene cuts. Snapping to scene cuts is a convention in anime fansubbing; some viewers find this lack of snapping to be distracting.

For many of the bonus features, the amount of black borders varied from scene to scene. It is time-consuming but possible to customize the cropping for each scene, ensuring a completely borderless video. Instead, the amount of cropping used was such that the thinnest border in the whole video clip is eliminated.

It’s possible to release more media files related to this anime series, such as DVD booklet scans and CD albums. There is no compelling reason to package everything together, so for now we’ll stick with just the anime and DVD bonus features.

The bonus features are not too interesting. Subbing the anime series was difficult enough; subbing the bonus features would just be extra work. Oh, and we don’t have a translator.

The project files package that was released is a snapshot at the culmination of the project. There are many intermediate snapshots too, but they are being kept private for the time being because they are less organized and because the changes between snapshots were not documented. If we were properly using a proper revision control system like SVN, then we would be in a better position to release the entire project files history.

To be explored

The following ideas are beyond what we can or are willing to experiment with at the moment.

Fansubbing involves heavy editing of plain text files over a period of time (days to months) by one or more people. Can revision control systems like SVN or Git be applied effectively to fansubbing – to keep work synchronized, to make changes visible and hold staff accountable for them, and to provide an archive of past states?

Instead of the simple, traditional animated fill karaoke provided by ASS, it would be nice to try some fancier effects. It seems that Adobe After Effects is one tool capable of this.

Fully automatic timing of dialogue lines may be difficult – for example, speakers need to be automatically differentiated, and President Aria’s voice is similar to speech but must be ignored. Automatic fine timing may be more feasible, only requiring the computer to find the exact start/end time when given a coarse human input that is accurate to within a few hundred milliseconds. Nonetheless, it still has the potential to save a lot of labor.

Errata

No one wants errors to end up in the published product. A lot of time was spent verifying the work, but errors inevitably arise.

In the bonus features, the Senpai/Kouhai 3 Ningumi omitted the word Interview. For example, Kouhai 3 Ningumi - Part 1 should actually be Kouhai 3 Ningumi Interview - Part 1.

It was discovered that 4×4 partitions for P frames hurt both quality and file size, and thus should have been excluded (although the effect is tiny). The option used was --partitions all (all = p8x8,p4x4,b8x8,i8x8,i4x4); the revised choice is p8x8,b8x8,i8x8,i4x4.

Frame 17727 (0-based) (~09:51) of episode 07 was discovered to be interlaced long after the release. It should have been deinterlaced.

Note that interlaced scene fades and interlaced text in opening/ending credits were not dealt with, as discussed in the section Steps not taken.

File information

Standard edition

Aria The Animation Blu Ray

Compact edition

Watch Aria The Animation Dub

Congratulations on reaching the end of the document. You were actually reading, weren’t you?

Aria The Animation Dub

Last modified: 2008-07-13-Sun
Created: 2007-09-11-Tue