完善资料让更多小伙伴认识你,还能领取20积分哦, 立即完善>
不确定是否有其他人使用了这些二进制LIB,但它们似乎没有准备好黄金时段。我买了两个都期待一个简单的现成的安装,但它不是这样的。(1)安装程序不与2.04.2的MP3 LIB:2手动添加LIBS和让它去也不起作用。我使用的是PIC32 MZ2064 DAH176,它有一个没有FPU的微APtiV核心,并且我怀疑这两种构建都没有正确地发挥作用。第3,微标签头与库定义不匹配,特别是Frp3GETStesteSex 4。该库看起来与驱动器中的原始MP3文件格格不入。给它喂食。我必须用勺子喂这个解码器吗?目前,我正在调用ITMP3StaseSeist= MP3DeCuffGestStaseSeIZE();MP3DeCuffStord= MaloC(MP3StaseSeI化);然后输入ITMP3DECODE(MP3DeXDeStRead,Ipple Buffor,InSimultand,Advand;Frand;OuttPuffor,Objsig);并且它大部分时间返回0。这是非常简化的,但是我把缓冲区的未使用部分移回开头,然后在它上面加上更多,等等。基本的LIB基本上没有记录。因此,在这一点上,这些看起来像是一个Wa$TE。
以上来自于百度翻译 以下为原文 Not sure if anyone else has used these binary libs, but they don't seem to be ready for prime time. I bought both expecting a simple ready to go setup, but it isn't to be so. #1 the installer doesn't mesh with 2.04. For the MP3 lib: #2 Manually adding the libs and getting it going doesn't work either. I am using the PIC32MZ2064DAH176 which has a microaptiv core without FPU, and I suspect that neither build plays properly. #3 The Microaptive header doesn't match the library definitions, particularly for MP3GetStateSize #4 The library seems geared to a pristine mp3 file from a drive, and I want to stream to it. Do I have to spoon feed this decoder? At present I am calling it Mp3StateSize = MP3DecoderGetStateSize(); Mp3decoderState = malloc(Mp3StateSize); Then feeding it MP3Decode(Mp3decoderState, inputBuffer, inSize, &used, &format, outputBuffer, &outSize)); and it returns 0 most of the time. This is very simplified, but I am moving the unused portion of the buffer back to the start before adding more on top of it, etc. The underlying libs are basically not documented. So at this point these seems like a wa$te. |
|
相关推荐
17个回答
|
|
我正在尝试MP3的东西,因为我ththHTTP://www. McCHIP.COM/FUMMS/FUNDSPE/1014506不确定它会走多远…
以上来自于百度翻译 以下为原文 I am trying MP3 stuff too, as per my thread http://www.microchip.com/forums/FindPost/1014506 Not sure how far it will go... |
|
|
|
你在易趣网上购买的Wave/mp3有很多芯片/模块,MPU的作用非常小,为MP3节省了大量的代码。
以上来自于百度翻译 以下为原文 There are chips/modules you buy on ebay that do wav/mp3. The mpu has do to very little. Saves a hell of a lot of coding for mp3. |
|
|
|
|
|
|
|
|
|
|
|
事实上是这样!我的意思是,在PC上,当我移植到C++类时,我发现大约有2001左右的“MPG123”代码。我喜欢音乐变成数学和回来的想法…
以上来自于百度翻译 以下为原文 Actually it was! I mean, on a PC, as I ported to a C++ class that "mpg123" code I found around, somewhere around 2001 or so! I loved the idea of music turning into math and back... |
|
|
|
|
|
|
|
对不起,论坛似乎没有警觉。我发现我没有正确地为初学者提供解码器。代码是为演示而设置的,所以文档不够好。代码每解码一个OP只使用一个帧,并且头文件也不完全正确。MyApTIV代码也不确定,我用MX Bug代替它工作。原则上,我不想把编码和解码交给另一个半烘焙的芯片。对于初学者来说,没有任何全双工编码器和解码器IC可用,第二,新的DA系列PIC32超过足够的能力。
以上来自于百度翻译 以下为原文 Sorry, forum doesn't seem to alert. I found I wasn't feeding the decoder correctly for starters. The code is set up for the demo, so the docs aren't good enough. The code only uses one frame per decode op, and the header docs aren't totally right either. Not sure about the microaptiv code either, I got it to work using the MX build instead. As a matter of principle, I didn't feel like handing the encoding and decoding off to another half baked chip. For starters, there aren't any full duplex encoder and decoder IC's available, and second, the new DA series pic32 are more than capable enough. |
|
|
|
埃里克,我在一些产品中使用过这些。至少在我的经验中,首先要注意的是,这些该死的东西像疯狂一样堆放在一起。也有一些最近的变化,我从微芯片直接得到了一些其他问题,在一些奇怪的情况下,我们不依赖于缓冲器的大小和一些汇编程序的数学源代码。我现在的一个应用程序与AAC的堆栈是32 K。谢谢,戈登。
以上来自于百度翻译 以下为原文 Eric, I have used these in a couple of products. The first thing to note at least in my experience is that the damn things eat stack like crazy. There were also some recent changes that I got direct from Microchips for some other issues which in weird circumstances would not work for us depending on size of buffers and some of the assembler math source code. My stack currently for one application with the AAC is 32K. Thanks, Gordon |
|
|
|
谢谢你提供的信息。对我来说,最麻烦的是为什么格式化。在我的测试中,SAMPLEATER在22050点流中显示了16000。在给解码器之前,你必须做任何消毒或帧定位吗?
以上来自于百度翻译 以下为原文 Thanks for that info. The most troubling thing to me currently is why the format.samplerate showed 16000 for a 22050 stream at one point in my testing. Did you have to do any sanitizing or frame locates before feeding the decoder? |
|
|
|
根据我的经验,所需要的是“帧头”,它通常是一个由11位设置为1的双字,其余的被检查(至少在我研究和使用的库中):
以上来自于百度翻译 以下为原文 From my experience, all that is needed is "frame header" that is usually a double-word made of 11 bits set to 1 and the rest is checked (at least in the library I studied and used) like this: if( /* first 11 bits are set to 1 for frame sync */ ((head & 0xffe00000) != 0xffe00000) || /* layer: 01,10,11 is 1,2,3; 00 is reserved */ (!((head>>17)&3)) || /* 1111 means bad bitrate */ (((head>>12)&0xf) == 0xf) || /* 0000 means free format... */ (((head>>12)&0xf) == 0x0) || /* sampling freq: 11 is reserved */ (((head>>10)&0x3) == 0x3 ) /* here used to be a mpeg 2.5 check... re-enabled 2.5 decoding due to lack of evidence that it is really not good */ ) { return FALSE; } |
|
|
|
埃里克,我所做的一切都是44.1K的输出,所以我甚至从来没有检查过。早先有一连串的内存泄漏,那些被修复了。堆栈的事情造成了最大的问题,因为我没有预料到这么大的大小。最近,我们有一个iOS和流死的问题,但他们也有一个处理。谢谢,戈登。
以上来自于百度翻译 以下为原文 Erik, Everything I was doing was 44.1K output, so I never even checked that. Early on there were a bunch of memory leaks, those got fixed. The stack thing caused the biggest problem because I did not anticipate that large of a size. Recently we had an issue with IOS and the stream going dead but they have a handle on that as well. Thanks, Gordon |
|
|
|
大家好,我们还没有更新网络上的解码器来发布两个版本。所以我们做的测试做的工作,编译和运行的盒子。所以我猜你们中的一些人都有比较旧的版本。给我发一个关于你购买的信息,我们可以先确定你有最新的下载。抱歉给你带来不便,但至少我们可以从一个共同点开始。
以上来自于百度翻译 以下为原文 Hi all, We have not updated the decoders on the web for a couple of releases. So the ones we have done our testing with do work, and compile and run out of the box. So my guess is that some of you all have older versions. Send me a PM with info about when you purchased, and we can start by making sure you have the latest download. Sorry for the inconvenience, but at least we can start at a common point. |
|
|
|
更多的更新。我通过Microchip直接查看了Web上的内容。我发现这确实是在内部更新的,但最近没有发布。所以和声V2.04有一个错配。我已经纠正了这个问题,并把所有的包裹都贴上了正确的更新版本。特别是在MP3中,这是不同的,因为我们在那里增加了特性。我建议任何购买工具(用于MPLAB和谐)的Microchip直接下载新版本。随着您的原始购买,您可以获得免费升级。如果你有任何问题,PM我和我会帮助你。
以上来自于百度翻译 以下为原文 More update. I reviewed what was up on the web through microchip Direct. I found that indeed this has been updated internally, but not posted recently. So the Harmony v2.04 had a mismatch. I have corrected this, and posted all packages with their correct updated versions. In MP3 in particular this makes a difference, since we have added features there. I advise any that have purchased the tool (for MPLAB Harmony) to go to microchip Direct and download the new version. With your original purchase you can get the upgrade for free. If you have any problem with this, PM me and I will help you out. |
|
|
|
好的,也许这会修正我所看到的错误的抽样问题。你能给最好的版本建议使用非FPU MaPaTIVDA系列吗?
以上来自于百度翻译 以下为原文 Good deal, perhaps this will fix that wrong samplerate issue I saw. Can you advise on the best version to use with the non fpu microaptiv DA series? |
|
|
|
HI FrESENEN MX构建版本在DA上工作,它被称为MP3LCubastCuffDr.Mx.a.该版本具有新的特性,如MPEG2和MPEG2.5解码支持。
以上来自于百度翻译 以下为原文 Hi friesen the MX build version works on DA, which is called MP3_Compact_Decoder_mx.a. This version has new features, such as MPEG2 and MPEG2.5 decoding support. |
|
|
|
好的,我更新了这些,现在我从AAC-LIB集成开始。为什么LIBS有更好的文档?对于LIBS联机的PDF文档是1.11,显然是用于测试应用程序。为什么没有合适的DOC和核心解码器?此外,我需要给解码器提供一个没有ADTS头的流。这能奏效吗?如何检索解码器中的采样率信息?函数描述只表示Untu8*Adts,对于任何东西都没有结构定义。我能为信息检索WaveHead,还是会用连续的流做一些有趣的事情?
以上来自于百度翻译 以下为原文 Ok, I updated those, and now I am starting on the AAC lib integration. Why are there better docs for the libs? The pdf documentation for the libs online is for 1.11, and apparently is for the test app. Why aren't there proper docs and headers for the core decoder? Also, I need to feed the decoder a stream without adts headers. Is this going to work? How can I retrieve the sample rate info from the decoder? The function description only says uint_8 * adts, there is no struct definition for anything. Can I retrieve a wavheader for info, or will that do something funny with a continuing stream? |
|
|
|
解码的立体声音频是交织的,但超长是所有解码音频的长度,而不是解码的样本。因此,在这个例子中,一个完整的解码帧是2304的长度,但是实际的立体样本是1152个,或者1152×2个通道。没有任何文档可以在任何地方解释这个问题。
以上来自于百度翻译 以下为原文 Some more notes on the premium mp3 decoder. Decoded stereo audio is interleaved, but outsize is length of all decoded audio, rather than decoded samples. So a full decoded frame is 2304 in length in this example, but the actual stereo samples are 1152, or 1152*2 channels. There are no docs to explain this anywhere. |
|
|
|
只有小组成员才能发言,加入小组>>
5243 浏览 9 评论
2035 浏览 8 评论
1955 浏览 10 评论
请问是否能把一个ADC值转换成两个字节用来设置PWM占空比?
3208 浏览 3 评论
请问电源和晶体值之间有什么关系吗?PIC在正常条件下运行4MHz需要多少电压?
2258 浏览 5 评论
779浏览 1评论
672浏览 1评论
有偿咨询,关于MPLAB X IPE烧录PIC32MX所遇到的问题
598浏览 1评论
PIC Kit3出现目标设备ID(00000000)与预期的设备ID(02c20000)不匹配。是什么原因
680浏览 0评论
578浏览 0评论
小黑屋| 手机版| Archiver| 电子发烧友 ( 湘ICP备2023018690号 )
GMT+8, 2024-12-24 12:50 , Processed in 1.757761 second(s), Total 110, Slave 94 queries .
Powered by 电子发烧友网
© 2015 bbs.elecfans.com
关注我们的微信
下载发烧友APP
电子发烧友观察
版权所有 © 湖南华秋数字科技有限公司
电子发烧友 (电路图) 湘公网安备 43011202000918 号 电信与信息服务业务经营许可证:合字B2-20210191 工商网监 湘ICP备2023018690号