multiple channel audio data and wave files.htm

WAV文件的格式官方宝典

      files. It was deemed impossible to define a default multi-channel speaker 
      configuration for <B>WAVE_FORMAT_PCM</B> without causing functionality to 
      break in one of these file types. Therefore, a channel-to-speaker 
      specification for <B>WAVE_FORMAT_PCM</B> will continue to be undocumented 
      for the case of <B>nChannels</B> &gt; 2.</P>
      <H3>Default Channel Ordering</H3>
      <P>The way to deterministically link channel numbers to speaker locations, 
      thus providing consistency among multiple channel audio files, is to 
      define the order in which the channels are laid out in the audio file. 
      Several external standards define parts of the following master channel 
      layout:</P>
      <TABLE class=numberedList cellSpacing=0 cellPadding=0 border=0>
        <TBODY>
        <TR vAlign=top>
          <TD class=listNumber noWrap align=right>
            <P>1.</P></TD>
          <TD>
            <P>Front Left - FL</P></TD></TR>
        <TR vAlign=top>
          <TD class=listNumber noWrap align=right>
            <P>2.</P></TD>
          <TD>
            <P>Front Right - FR</P></TD></TR>
        <TR vAlign=top>
          <TD class=listNumber noWrap align=right>
            <P>3.</P></TD>
          <TD>
            <P>Front Center - FC</P></TD></TR>
        <TR vAlign=top>
          <TD class=listNumber noWrap align=right>
            <P>4.</P></TD>
          <TD>
            <P>Low Frequency - LF</P></TD></TR>
        <TR vAlign=top>
          <TD class=listNumber noWrap align=right>
            <P>5.</P></TD>
          <TD>
            <P>Back Left - BL</P></TD></TR>
        <TR vAlign=top>
          <TD class=listNumber noWrap align=right>
            <P>6.</P></TD>
          <TD>
            <P>Back Right - BR</P></TD></TR>
        <TR vAlign=top>
          <TD class=listNumber noWrap align=right>
            <P>7.</P></TD>
          <TD>
            <P>Front Left of Center - FLC</P></TD></TR>
        <TR vAlign=top>
          <TD class=listNumber noWrap align=right>
            <P>8.</P></TD>
          <TD>
            <P>Front Right of Center - FRC</P></TD></TR>
        <TR vAlign=top>
          <TD class=listNumber noWrap align=right>
            <P>9.</P></TD>
          <TD>
            <P>Back Center - BC</P></TD></TR>
        <TR vAlign=top>
          <TD class=listNumber noWrap align=right>
            <P>10.</P></TD>
          <TD>
            <P>Side Left - SL</P></TD></TR>
        <TR vAlign=top>
          <TD class=listNumber noWrap align=right>
            <P>11.</P></TD>
          <TD>
            <P>Side Right - SR</P></TD></TR>
        <TR vAlign=top>
          <TD class=listNumber noWrap align=right>
            <P>12.</P></TD>
          <TD>
            <P>Top Center - TC</P></TD></TR>
        <TR vAlign=top>
          <TD class=listNumber noWrap align=right>
            <P>13.</P></TD>
          <TD>
            <P>Top Front Left - TFL</P></TD></TR>
        <TR vAlign=top>
          <TD class=listNumber noWrap align=right>
            <P>14.</P></TD>
          <TD>
            <P>Top Front Center - TFC</P></TD></TR>
        <TR vAlign=top>
          <TD class=listNumber noWrap align=right>
            <P>15.</P></TD>
          <TD>
            <P>Top Front Right - TFR</P></TD></TR>
        <TR vAlign=top>
          <TD class=listNumber noWrap align=right>
            <P>16.</P></TD>
          <TD>
            <P>Top Back Left - TBL</P></TD></TR>
        <TR vAlign=top>
          <TD class=listNumber noWrap align=right>
            <P>17.</P></TD>
          <TD>
            <P>Top Back Center - TBC</P></TD></TR>
        <TR vAlign=top>
          <TD class=listNumber noWrap align=right>
            <P>18.</P></TD>
          <TD>
            <P>Top Back Right - TBR</P></TD></TR></TBODY></TABLE>
      <P>The channels in the interleaved stream corresponding to these spatial 
      positions must appear in the order specified above. This holds true even 
      in the case of a non-contiguous subset of channels. For example, if a 
      stream contains left, bass enhance and right, then channel 1 is left, 
      channel 2 is right, and channel 3 is bass enhance. This enables the 
      linkage of multi-channel streams to well-defined multi-speaker 
      configurations.</P>
      <P><B>Warning:</B> Content intended for the Low Frequency channel may not 
      be rendered on the speaker that the data is sent to. This is because there 
      is no way to guarantee the frequency range of the low frequency speaker in 
      a user's system. For this reason, a speaker that is receiving low 
      frequency audio might filter the frequencies that it cannot handle.</P>
      <DIV style="MARGIN-TOP: 3px; MARGIN-BOTTOM: 10px"><A 
      href="http://www.microsoft.com/whdc/hwdev/tech/audio/multichaud.mspx#top"><IMG 
      height=9 alt="Top of page" 
      src="Multiple Channel Audio Data and WAVE Files.files/arrow_px_up.gif" 
      width=7 border=0></A><A class=topOfPage 
      href="http://www.microsoft.com/whdc/hwdev/tech/audio/multichaud.mspx#top">Top 
      of page</A></DIV><A name=#XSLTsection125121120120></A>
      <H2>Representing High-Resolution Audio </H2>
      <H3>Ambiguity within WAVE_FORMAT_PCM</H3>
      <P>It is important to enumerate the assumptions made by 
      <B>WAVE_FORMAT_PCM</B>. One assumption is that the <B>nBlockAlign</B> 
      field contains exactly one set of samples (one block). Also, each sample 
      must be byte-aligned within the block. The byte-aligned space that each 
      sample consumes is called the sample's "container." Additionally, there 
      cannot be any extra space between the end of the last sample container and 
      the actual end of the block. In other words, <B>nBlockAlign</B> must be an 
      integer multiple of <B>nChannels</B>, and that multiple is considered the 
      "container size" for the sample.</P>
      <P>Although it was not necessarily the original intent, some entities 
      treat <B>wBitsPerSample</B> as equivalent to the container size. In other 
      words, the number of bits of valid data was assumed equal to the size of 
      the container. This assumption held well for 8-bit and 16-bit audio. With 
      the coming of high fidelity audio equipment for render and capture, 20-bit 
      and 24-bit streams emerged. Specifying the actual bit resolution of the 
      samples in a byte-aligned stream became important; the accepted 
      interpretation of <B>wBitsPerSample</B> started to unravel, because its 
      value is not rigidly enforced by all software. In many cases, it continued 
      to be the container size. In other cases, however, the field was used to 
      indicate the bits of actual valid data, while the container size was 
      inferred from <B>nBlockAlign</B> and <B>nChannels</B> (an example would be 
      <B>wBitsPerSample</B> = 20, <B>nBlockAlign</B> = 8, <B>nChannels</B> = 
      2).</P>
      <P>It was not considered feasible to resolve this issue either way without 
      causing many legacy problems for <B>WAVE_FORMAT_PCM</B>, so the new 
      formats include an additional field that clarifies this ambiguity. The new 
      formats are intended to handle those cases in which the actual bits of 
      precision are not equal to the container size, as well as all cases in 
      which the container size is greater than sixteen bits.</P>
      <P>It is not the purpose of this article to resolve the ambiguity of 
      <B>wBitsPerSample</B> for <B>WAVE_FORMAT_PCM</B>. However, this same field 
      will be used unambiguously in the new formats.</P>
      <H3>Specifying the Actual Bit Depth</H3>
      <P>In the new formats, <B>wBitsPerSample</B> is strictly defined as the 
      container size. Containers must be byte-aligned, so <B>wBitsPerSample</B> 
      must be a multiple of 8. A new field conveys exactly how many of those 
      bits contain actual data, regardless of the size of the container. As 
      audio goes through a processing system, the number of valid data bits per 
      sample can change. Each sample is justified to the most significant bit, 
      so it is easy to add additional precision on the bottom of a sample, or 
      dither to a lower precision. The container size can be manipulated 
      independently without implying any change in the data precision. A 24-bit 
      stream in 32-bit containers (for efficient processing) can safely be 
      transferred into 24-bit containers (for efficient storage space) without 
      any data loss.</P>
      <DIV style="MARGIN-TOP: 3px; MARGIN-BOTTOM: 10px"><A 
      href="http://www.microsoft.com/whdc/hwdev/tech/audio/multichaud.mspx#top"><IMG 
      height=9 alt="Top of page" 
      src="Multiple Channel Audio Data and WAVE Files.files/arrow_px_up.gif" 
      width=7 border=0></A><A class=topOfPage 
      href="http://www.microsoft.com/whdc/hwdev/tech/audio/multichaud.mspx#top">Top 
      of page</A></DIV><A name=#XSLTsection126121120120></A>
      <H2>Using WAVE_FORMAT_EXTENSIBLE </H2>
      <P>To solve the problem of multiple channel ordering and high precision 
      data, Microsoft has defined the new structure 
      <B>WAVE_FORMAT_EXTENSIBLE</B>. This new structure not only has the benefit 
      of solving these problems, it also provides a mechanism for 
      self-registering new data formats. The <B>SubFormat</B> field is set to 
      the GUID that specifies the type of data described by the 
      <B>WAVE_FORMAT_EXTENSIBLE</B> structure.</P>
      <DIV style="MARGIN-TOP: 3px; MARGIN-BOTTOM: 10px"><A 
      href="http://www.microsoft.com/whdc/hwdev/tech/audio/multichaud.mspx#top"><IMG 
      height=9 alt="Top of page" 
      src="Multiple Channel Audio Data and WAVE Files.files/arrow_px_up.gif" 
      width=7 border=0></A><A class=topOfPage 
      href="http://www.microsoft.com/whdc/hwdev/tech/audio/multichaud.mspx#top">Top 
      of page</A></DIV><A name=#XSLTsection127121120120></A>
      <H2>Definition of WAVE_FORMAT_EXTENSIBLE </H2>
      <P>The definition (in MMREG.H and KSMEDIA.H) of 
      <B>WAVE_FORMAT_EXTENSIBLE</B> is included below:</P><PRE class=codeSample>typedef struct {
    WAVEFORMATEX  Format;
    union {
        WORD wValidBitsPerSample; /* bits of precision */
        WORD wSamplesPerBlock;   /* valid if wBitsPerSample==0 */
        WORD wReserved;         /* If neither applies, set to zero. */
    } Samples;
    DWORD    dwChannelMask; /* which channels are present in stream */
    GUID            SubFormat;
} WAVEFORMATEXTENSIBLE, *PWAVEFORMATEXTENSIBLE;
</PRE>
      <P><BR><B>Definition of WAVEFORMATPCMEX</B><BR>The wave format 
      <B>WAVEFORMATEX</B> is too under-specified to be used for high bit-depth 
      samples or multiple channel streams. For cases in which the spatial 
      locations of the channels are linked to the standard speaker locations, 
      <B>WAVEFORMATPCMEX</B> is appropriate. For the <B>WAVEFORMATEXTENSIBLE</B>