Chinaunix首页 | 论坛 | 博客
  • 博客访问: 385784
  • 博文数量: 61
  • 博客积分: 4650
  • 博客等级: 上校
  • 技术积分: 786
  • 用 户 组: 普通用户
  • 注册时间: 2008-11-11 21:07
个人简介

少抱怨,多实干;

文章分类

全部博文(61)

文章存档

2017年(1)

2016年(13)

2015年(1)

2013年(2)

2011年(1)

2010年(3)

2009年(23)

2008年(17)

我的朋友

分类: LINUX

2008-11-23 16:01:10

一.概述:

MP3文件是由帧(frame)构成的,帧是MP3文件最小的组成单位。MP3的全称应为MPEG1 Layer-3音频文件,MPEG(Moving Picture Experts Group)在汉语中译为活动图像专家组,特指活动影音压缩标准,MPEG音频文件是MPEG1标准中的声音部分,也叫MPEG音频层,它根据压缩质量和编码复杂程度划分为三层,即Layer-1Layer2Layer3,且分别对应MP1MP2MP3这三种声音文件,并根据不同的用途,使用不同层次的编码。MPEG音频编码的层次越高,编码器越复杂,压缩率也越高,MP1MP2的压缩率分别为4161-81,而MP3的压缩率则高达101-121,也就是说,一分钟CD音质的音乐,未经压缩需要10MB的存储空间,而经过MP3压缩编码后只有1MB左右。不过MP3对音频信号采用的是有损压缩方式,为了降低声音失真度,MP3采取了感官编码技术,即编码时先对音频文件进行频谱分析,然后用过滤器滤掉噪音电平,接着通过量化的方式将剩下的每一位打散排列,最后形成具有较高压缩比的MP3文件,并使压缩后的文件在回放时能够达到比较接近原音源的声音效果。

 

二.整个MP3文件结构:

MP3文件大体分为三部分:TAG_V2(ID3V2)Frame,  TAG_V1(ID3V1)

ID3V2

包含了作者,作曲,专辑等信息,长度不固定,扩展了ID3V1的信息量。

Frame

.

.

.

Frame

一系列的帧,个数由文件大小和帧长决定

每个FRAME的长度可能不固定,也可能固定,由位率bitrate决定

每个FRAME又分为帧头和数据实体两部分

帧头记录了mp3的位率,采样率,版本等信息,每个帧之间相互独立

ID3V1

包含了作者,作曲,专辑等信息,长度为128BYTE

 

 









三.
MP3FRAME格式:

每个FRAME都有一个帧头FRAMEHEADER,长度是4BYTE32bit,帧头后面可能有两个字节的CRC校验,这两个字节的是否存在决定于FRAMEHEADER信息的第16bit,为0则帧头后面无校验,为1则有校验,校验值长度为2个字节,紧跟在FRAMEHEADER后面,接着就是帧的实体数据了,格式如下:

 

FRAMEHEADER

CRCfree

MAIN_DATA

4 BYTE

0 OR 2 BYTE

长度由帧头计算得出

 

1.帧头FRAMEHEADER格式如下:

AAAAAAAA AAABBCCD EEEEFFGH IIJJKLMM

13个帧头字符的含义如下:

Sign

Length
(bits)

Position
(bits)

Description

A

11

(31-21)

Frame sync (all bits set)

B

2

(20,19)

MPEG Audio version
00 - MPEG Version 2.5
01 - reserved
10 - MPEG Version 2
11 - MPEG Version 1

C

2

(18,17)

Layer description
00 - reserved
01 - Layer III
10 - Layer II
11 - Layer I

D

1

(16)

Protection bit
0 - Protected by CRC (16bit crc follows header)
1 - Not protected

E

4

(15,12)

Bitrate index

bits

V1,L1

V1,L2

V1,L3

V2,L1

V2,L2

V2,L3

0000

free

free

free

free

free

free

0001

32

32

32

32

32

8 (8)

0010

64

48

40

64

48

16 (16)

0011

96

56

48

96

56

24 (24)

0100

128

64

56

128

64

32 (32)

0101

160

80

64

160

80

64 (40)

0110

192

96

80

192

96

80 (48)

0111

224

112

96

224

112

56 (56)

1000

256

128

112

256

128

64 (64)

1001

288

160

128

288

160

128 (80)

1010

320

192

160

320

192

160 (96)

1011

352

224

192

352

224

112 (112)

1100

384

256

224

384

256

128 (128)

1101

416

320

256

416

320

256 (144)

1110

448

384

320

448

384

320 (160)

1111

bad

bad

bad

bad

bad

bad

NOTES: All values are in kbps
V1 - MPEG Version 1
V2 - MPEG Version 2 and Version 2.5
L1 - Layer I
L2 - Layer II
L3 - Layer III
"free" means variable bitrate.
"bad" means that this is not an allowed value

The values in parentheses are from different sources which claim that those values are valid for V2,L2 and V2,L3. If anyone can confirm please let me know.

F

2

(11,10)

Sampling rate frequency index (values are in Hz)

bits

MPEG1

MPEG2

MPEG2.5

00

44100

22050

11025

01

48000

24000

12000

10

32000

16000

8000

11

reserv.

reserv.

reserv.

G

1

(9)

Padding bit
0 - frame is not padded
1 - frame is padded with one extra bit

H

1

(8)

Private bit (unknown purpose)

I

2

(7,6)

Channel Mode
00 - Stereo
01 - Joint stereo (Stereo)
10 - Dual channel (Stereo)
11 - Single channel (Mono)

J

2

(5,4)

Mode extension (Only if Joint stereo)

value

Intensity stereo

MS stereo

00

off

off

01

on

off

10

off

on

11

on

on

K

1

(3)

Copyright
0 - Audio is not copyrighted
1 - Audio is copyrighted

L

1

(2)

Original
0 - Copy of original media
1 - Original media

M

2

(1,0)

Emphasis
00 - none
01 - 50/15 ms
10 - reserved
11 - CCIT J.17

1)每帧的播放时间:无论帧长是多少,每帧的播放时间都是26ms

2)数据帧大小:

FrameSize = (((MpegVersion == MPEG1 ? 144 : 72) * Bitrate) / SamplingRate) + PaddingBit

例如:  Bitrate = 128000, a SamplingRate =44100, and PaddingBit = 1

 FrameSize = (144 * 128000) / 44100 + 1  = 417 bytes

 

2MAIN_DATA

MAIN_DATA部分长度是否变化决定于FRAMEHEADERbitrate是否变化,一MP3歌曲,它有三个版本:96Kbps(96千比特位每秒)、128Kbps和192Kbps。Kbps(比特位速率),表明了音乐每秒的数据量,Kbps值越高,音质越好,文件也越大,MP3标准规定,不变的bitrateMP3文件称作CBR,大多数MP3文件都是CBR的,而变化的bitrateMP3文件称作VBR,每个FRAME的长度都可能是变化的。下面是CBRVBR的不同点:

 

1CBR:固定位率的FRAME的大小也是固定的(公式如上所述),只要知道文件总长度,和帧长即可由播放每帧需26ms计算得出mp3播放的总时间,也可通过计数帧的个数控制快进、快退慢放等操作。

 

2VBRVBRXING公司推出的算法,所以在MP3FRAME里会有“XING"这个关键字(现在很多流行的小软件也可以进行VBR压缩,它们是否遵守这个约定,那就不得而知了),它存放在MP3文件中的第一个有效FRAME里,它标识了这个MP3文件是VBR的。同时第一个FRAME里存放了MP3文件的FRAME的总个数,这就很容易获得了播放总时间,同时还有100个字节存放了播放总时间的100个时间分段的FRAMEINDEX,假设4分钟的MP3歌曲,240S,分成100段,每两个相邻INDEX的时间差就是2.4S,所以通过这个INDEX,只要前后处理少数的FRAME,就能快速找出我们需要快进的FRAME头,可参考下文

This system was created to minimize file lengths and to preserve sound quality.
Higher frequencies generally needs more space for encoding (thats why many codecs cut all frequencies above cca 16kHz) and lower tones requires less. So if some part of song doesnt consist of higher tones then using eg. 192kbps is wasting of space. It should be enough to use only eg. 96kbps.
And it is the principle of VBR. Codec looks over frame and then choose bitrate suitable for its sound quality.

It sounds perfect but it brings some problems:
If you want to jump over 2 minutes in song, it is not a problem with CBR because you are able simply count amount of Bytes which is necessary to skip. But it is impossible with VBR. Frame lengths should be arbitrary so you have to either go frame by frame and counts (time consuming and very unpractical) or use another mechanism for approximate count.
If you want to cut 5 minutes from the middle of VBR file (all we know CDs where last song takes 10 minutes but 5 minutes is a pure silence, HELL!) problems are the same.
Result? VBR files are more difficult for controlling and adjusting. And I dont like feeling that sound quality changes in every moment. And AFAIK many codecs have problems with creation VBR in good quality.
Personally I cant see any reason why to use VBR - I dont give a fuck if size of one CD in MP3 is 55 MB with CBR or 51 MB with VBR. But everybody has a different taste... some people prefer VBR.

VBR File Structure

is the same as for CBR. But the first frame doesnt contain audio data and it is used for special information about VBR file.

Structure of the first frame:

Byte      

  

Content

0-3

 

Standard audio frame header (as descripted above). Mostly it contains values FF FB 30 4C, from which you can count FrameLen = 156 Bytes. And thats exactly enough space for storing VBR info.
This header contains some important information valid for the whole file:
- MPEG (MPEG1 or MPEG2)
- SAMPLING rate frequency index
- CHANNEL (JointStereo etc.)

  

 

 

4-x

 

Not used till string "Xing" (58 69 6E 67). This string is used as a main VBR file identifier. If it is not found, file is supposed to be CBR. This string can be placed at different locations according to values of MPEG and CHANNEL (ya, these from a few lines upwards):

36-39

 

"Xing" for MPEG1 and CHANNEL != mono (mostly used)

21-24

 

"Xing" for MPEG1 and CHANNEL == mono

21-24

 

"Xing" for MPEG2 and CHANNEL != mono

13-16

 

"Xing" for MPEG2 and CHANNEL == mono

  

 

 

 

 

After "Xing" string there are placed flags, number of frames in file and a size of file in Bytes. Each of these items has 4 Bytes and it is stored as 'int' number in memory. The first is the most significant Byte and the last is the least.


Following schema is for MPEG1 and CHANNEL != mono:

40-43

 

Flags

Value

  

Name

  

Description

00 00 00 01

 

Frames Flag

 

set if value for number of frames in file is stored

00 00 00 02

 

Bytes Flag

 

set if value for filesize in Bytes is stored

00 00 00 04

 

TOC Flag

 

set if values for TOC (see below) are stored

00 00 00 08

 

VBR Scale Flag

 

set if values for VBR scale are stored

All these values can be stored simultaneously.

  

 

 

44-47

 

Frames
Number of frames in file (including the first info one)

  

 

 

48-51

 

Bytes
File length in Bytes

  

 

 

52-151

 

TOC (Table of Contents)
Contains of 100 indexes (one Byte length) for easier lookup in file. Approximately solves problem with moving inside file.
Each Byte has a value according this formula:
(TOC[i] / 256) * fileLenInBytes
So if song lasts eg. 240 sec. and you want to jump to 60. sec. (and file is 5 000 000 Bytes length) you can use:
TOC[(60/240)*100] = TOC[25]
and corresponding Byte in file is then approximately at:
(TOC[25]/256) * 5000000

If you want to trim VBR file you should also reconstruct Frames, Bytes and TOC properly.

  

 

 

152-155

 

VBR Scale
I dont know exactly system of storing of this values but this item probably doesnt have deeper meaning.

 

四.ID3v1
         ID3V1
比较简单,它是存放在MP3文件的末尾,用16进制的编辑器打开一个MP3文件,查看其末尾的128个顺序存放字节,数据结构定义如下:
 typedef struct tagID3V1

{

char Header[3];    /*标签头必须是"TAG"否则认为没有标签*/
char Title[30];    /*
标题*/
char Artist[30];   /*
作者*/
char Album[30];    /*
专集*/
char Year[4];    /*
出品年代*/
char Comment[28];   /*
备注*/

char reserve;      /*保留*/

char track;; /*音轨*/
char Genre;    /*
类型*/

}ID3V1,*pID3V1;

 

         ID3V1的各项信息都是顺序存放,没有任何标识将其分开,比如标题信息不足30个字节,则使用'\0'补足,否则将造成信息错误。Genre使用原码表示,对照表如下:
 /* Standard genres */
 0="Blues";
 1="ClassicRock";
 2="Country";
 3="Dance";
 4="Disco";
 5="Funk";
 6="Grunge";
 7="Hip-Hop";
 8="Jazz";
 9="Metal";
 10="NewAge";
 11="Oldies";
 12="Other";
 13="Pop";
 14="R&B";
 15="Rap";
 16="Reggae";
 17="Rock";
 18="Techno";
 19="Industrial";
 20="Alternative";
 21="Ska";
 22="DeathMetal";
 23="Pranks";
 24="Soundtrack";
 25="Euro-Techno";
 26="Ambient";
 27="Trip-Hop";
 28="Vocal";
 29="Jazz+Funk";
 30="Fusion";
 31="Trance";
 32="Classical";
 33="Instrumental";
 34="Acid";
 35="House";
 36="Game";
 37="SoundClip";
 38="Gospel";
 39="Noise";
 40="AlternRock";
 41="Bass";
 42="Soul";
 43="Punk";
 44="Space";
 45="Meditative";
 46="InstrumentalPop";
 47="InstrumentalRock";
 48="Ethnic";
 49="Gothic";
 50="Darkwave";
 51="Techno-Industrial";
 52="Electronic";
 53="Pop-Folk";
 54="Eurodance";
 55="Dream";
 56="SouthernRock";
 57="Comedy";
 58="Cult";
 59="Gangsta";
 60="Top40";
 61="ChristianRap";
 62="Pop/Funk";
 63="Jungle";
 64="NativeAmerican";
 65="Cabaret";
 66="NewWave";
 67="Psychadelic";
 68="Rave";
 69="Showtunes";
 70="Trailer";
 71="Lo-Fi";
 72="Tribal";
 73="AcidPunk";
 74="AcidJazz";
 75="Polka";
 76="Retro";
 77="Musical";
 78="Rock&Roll";
 79="HardRock";
 /* Extended genres */
 80="Folk";
 81="Folk-Rock";
 82="NationalFolk";
 83="Swing";
 84="FastFusion";
 85="Bebob";
 86="Latin";
 87="Revival";
 88="Celtic";
 89="Bluegrass";
 90="Avantgarde";
 91="GothicRock";
 92="ProgessiveRock";
 93="PsychedelicRock";
 94="SymphonicRock";
 95="SlowRock";
 96="BigBand";
 97="Chorus";
 98="EasyListening";
 99="Acoustic";
 100="Humour";
 101="Speech";
 102="Chanson";
 103="Opera";
 104="ChamberMusic";
 105="Sonata";
 106="Symphony";
 107="BootyBass";
 108="Primus";
 109="PornGroove";
 110="Satire";
 111="SlowJam";
 112="Club";
 113="Tango";
 114="Samba";
 115="Folklore";
 116="Ballad";
 117="PowerBallad";
 118="RhythmicSoul";
 119="Freestyle";
 120="Duet";
 121="PunkRock";
 122="DrumSolo";
 123="Acapella";
 124="Euro-House";
 125="DanceHall";
 126="Goa";
 127="Drum&Bass";
 128="Club-House";
 129="Hardcore";
 130="Terror";
 131="Indie";
 132="BritPop";
 133="Negerpunk";
 134="PolskPunk";
 135="Beat";
 136="ChristianGangstaRap";
 137="HeavyMetal";
 138="BlackMetal";
 139="Crossover";
 140="ContemporaryChristian";
 141="ChristianRock";
 142="Merengue";
 143="Salsa";
 144="TrashMetal";
 145="Anime";
 146="JPop";
 147="Synthpop";


五.ID3V2
       ID3V2
到现在一共有4个版本,但流行的播放软件一般只支持第3版,既ID3v2.3。由于ID3V1记录在MP3文件的末尾,ID3V2就只好记录在MP3文件的首部了(如果有一天发布ID3V3,真不知道该记录在哪里)。也正是由于这个原因,对ID3V2的操作比ID3V1要慢。而且ID3V2结构比ID3V1的结构要复杂得多,但比前者全面且可以伸缩和扩展。
 
下面就介绍一下ID3V2.3
 
每个ID3V2.3的标签都一个标签头和若干个标签帧或一个扩展标签头组成。关于曲目的信息如标题、作者等都存放在不同的标签帧中,扩展标签头和标签帧并不是必要的,但每个标签至少要有一个标签帧。标签头和标签帧一起顺序存放在MP3文件的首部。
 
1
、标签头
在文件的首部顺序记录10个字节的ID3V2.3的头部。数据结构如下:
 char Header[3];    /*
必须为"ID3"否则认为标签不存在*/
 char Ver;    /*
版本号ID3V2.3就记录3*/
 char Revision;    /*
副版本号此版本记录为0*/
 char Flag;    /*
存放标志的字节,这个版本只定义了三位,稍后详细解说*/
 char Size[4];    /*
标签大小,包括标签头的10个字节和所有的标签帧的大小*/

 1).标志字节
 标志字节一般为0,定义如下:
 abc00000

 a -- 表示是否使用Unsynchronisation(这个单词不知道是什么意思,字典里也没有找到,一般不设置)
 b -- 表示是否有扩展头部,一般没有(至少Winamp没有记录),所以一般也不设置
 c -- 表示是否为测试标签(99.99%的标签都不是测试用的啦,所以一般也不设置)
 
 2).标签大小
 一共四个字节,但每个字节只用7位,最高位不使用恒为0。所以格式如下
 0xxxxxxx 0xxxxxxx 0xxxxxxx 0xxxxxxx
 计算大小时要将0去掉,得到一个28位的二进制数,就是标签大小(不懂为什么要这样做),计算公式如下:
 int total_size;
 total_size =   (Size[0]&0x7F)*0x200000
   +(Size[1]&0x7F)*0x400
   +(Size[2]&0x7F)*0x80
   +(Size[3]&0x7F)
 
2、标签帧
每个标签帧都有一个10个字节的帧头和至少一个字节的不固定长度的内容组成。它们也是顺序存放在文件中,和标签头和其他的标签帧也没有特殊的字符分隔。得到一个完整的帧的内容只有从帧头中的到内容大小后才能读出,读取时要注意大小,不要将其他帧的内容或帧头读入。

帧头的定义如下:
 char FrameID[4];  /*用四个字符标识一个帧,说明其内容,稍后有常用的标识对照表*/
 char Size[4];   /*帧内容的大小,不包括帧头,不得小于1*/ 
 char Flags[2];   /*存放标志,只定义了6位,稍后详细解说*/
 
 1).帧标识
 用四个字符标识一个帧,说明一个帧的内容含义,常用的对照如下:
 TIT2=标题 表示内容为这首歌的标题,下同
 TPE1=作者
 TALB=专集
 TRCK=音轨 格式:N/M  其中N为专集中的第N首,M为专集中共M首,N和M为ASCII码表示的数字
 TYER=年代 是用ASCII码表示的数字
 TCON=类型 直接用字符串表示
 COMM=备注 格式:"eng\0备注内容",其中eng表示备注所使用的自然语言

 2).大小
 这个可没有标签头的算法那么麻烦,每个字节的8位全用,格式如下
 xxxxxxxx xxxxxxxx xxxxxxxx xxxxxxxx
 算法如下:
 int FSize;
 FSize =  Size[0]*0x100000000
  +Size[1]*0x10000
  +Size[2]*0x100
  +Size[3];
  
 3).标志
 只定义了6位,另外的10位为0,但大部分的情况下16位都为0就可以了。格式如下:
 abc00000 ijk00000
 a -- 标签保护标志,设置时认为此帧作废
 b -- 文件保护标志,设置时认为此帧作废
 c -- 只读标志,设置时认为此帧不能修改(但我没有找到一个软件理会这个标志)
 i -- 压缩标志,设置时一个字节存放两个BCD码表示数字
 j -- 加密标志(没有见过哪个MP3文件的标签用了加密)
 k -- 组标志,设置时说明此帧和其他的某帧是一组
 
 值得一提的是winamp在保存和读取帧内容的时候会在内容前面加个'\0',并把这个字节计算在帧内容的大小中。
 
 详细的情况可以到
查询,对于ID3V1和ID3V2的读写,我用DELPHI写了两个类来实现,可以写信给我索取
 
附:帧标识的含义
4).   Declared ID3v2 frames

   The following frames are declared in this draft.

  AENC Audio encryption
  APIC Attached picture

  COMM Comments
  COMR Commercial frame

  ENCR Encryption method registration
  EQUA Equalization
  ETCO Event timing codes

  GEOB General encapsulated object
  GRID Group identification registration

  IPLS Involved people list

  LINK Linked information

  MCDI Music CD identifier
  MLLT MPEG location lookup table

  OWNE Ownership frame

  PRIV Private frame
  PCNT Play counter
  POPM Popularimeter
  POSS Position synchronisation frame

  RBUF Recommended buffer size
  RVAD Relative volume adjustment
  RVRB Reverb

  SYLT Synchronized lyric/text
  SYTC Synchronized tempo codes

  TALB Album/Movie/Show title
  TBPM BPM (beats per minute)
  TCOM Composer
  TCON Content type
  TCOP Copyright message
  TDAT Date
  TDLY Playlist delay
  TENC Encoded by
  TEXT Lyricist/Text writer
  TFLT File type
  TIME Time
  TIT1 Content group description
  TIT2 Title/songname/content description
  TIT3 Subtitle/Description refinement
  TKEY Initial key
  TLAN Language(s)
  TLEN Length
  TMED Media type
  TOAL Original album/movie/show title
  TOFN Original filename
  TOLY Original lyricist(s)/text writer(s)
  TOPE Original artist(s)/performer(s)
  TORY Original release year
  TOWN File owner/licensee
  TPE1 Lead performer(s)/Soloist(s)
  TPE2 Band/orchestra/accompaniment
  TPE3 Conductor/performer refinement
  TPE4 Interpreted, remixed, or otherwise modified by
  TPOS Part of a set
  TPUB Publisher
  TRCK Track number/Position in set
  TRDA Recording dates
  TRSN Internet radio station name
  TRSO Internet radio station owner
  TSIZ Size
  TSRC ISRC (international standard recording code)
  TSSE Software/Hardware and settings used for encoding
  TYER Year
  TXXX User defined text information frame

  UFID Unique file identifier
  USER Terms of use
  USLT Unsychronized lyric/text transcription

  WCOM Commercial information
  WCOP Copyright/Legal information
  WOAF Official audio file webpage
  WOAR Official artist/performer webpage
  WOAS Official audio source webpage
  WORS Official internet radio station homepage
  WPAY Payment
  WPUB Publishers official webpage
  WXXX User defined URL link frame

 
转自:
作者jmxuan发表于 2004年08月03日
阅读(555) | 评论(0) | 转发(0) |
给主人留下些什么吧!~~