Seeking Direction with MP3 Audiobook Conversion Project Options

[Ultra57]

First, let me start off by saying this forum is a gold mine of information. Users appear to be very knowledgeable and helpful. Thanks goes to both the users and moderators for creating such a valuable online resource. Second, this is a duplicate of a post in Off-Topic (a kind user suggested I repost here stating that I was Off-Topic in the Off-Topic area of the forum, and I would appreciate if a Moderator would delete my first post since I have added questions in this post).

I am here to research about MP3 codices for the purpose of audiobooks or voice recordings (I have over 6 TB of audiobook files needing conversion). I am looking for information regarding programs and codex information. Unfortunately, my experience level is limited to successfully installing (with the help of a step by step "just do this" tutorial) Exact Audio Copy (EAC) on four machines along with installation of the LAME encoder (also provided by the step by step) and a few settings within EAC that I really do not fully understand, but was able to tweak a bit with some success (CBR vs. VBR and ~ play with the bit rate settings).

For me, it is now the time to reach a better understanding of what these settings actually mean and what are my best choices. I read a little about Speex. I also read a very informative thread in regards to playback issues at levels <32 k by an actual user (vs. a well informed user guessing). Unfortunately, most of the posts leave me just wondering because my knowledge level is so low. I was hoping for some direction for my research within the site and the WIKKI to get me on my way. Questions I am looking to research here:

• What is Sampling Rate?
• What does that kbps rate mean to the outcome to my audiobook files and what a best starting point?
• Will EAC handle multiple MP3 codices?
• More information on EAC? FOUND IN THE FAQs and WIKI and still reading
• What does the MHtz Sampling Frequency refer to and what are it's ramification on sound quality in the area of speech?
• Essentially, how do the kbps and Mhtz settings come into play and what do they mean?
• Can I take the MP3PRO codex (reportedly very good for speech recordings but who knows for sure) from my copy of Nero and use it with EAC?
• Other than a critical ear, what evaluation programs are available to test resulting MP3 wave files and how do I interpret the test results?
• Reference is often made to a V0 or VX setting for the LAME encoder in the forum. What does this mean and is it part of the now mysterious set up of EAC? Found information on this in the WIKI, but still unclear on 0, 1, 2, ... and what these mean) Still reading.
• Decoder vs. Encoder vs. Other - I have assumed (which means I am probably wrong) the MP3 Decoder is used when you start with an MP3 file and a MPE Encoder is when you want to make an MP3 file. Other is just that right now. What other things are used/needed to create an MP3? Evidently, LAME is an other.
• I have attempted to read the Nero manual with all it's technical information on presets. I was curious about some of the terms that are obviously tied to old audio editing techniques (i.e., volume control vs. normalization, anti-ticking vs. anti-clicking vs. anti-hissing and their role in maybe eliminating the clicks from a scratched CD, etc.). Where can I read up on this type of information to make more informed decisions on the best tool and settings for these correction techniques.
• What is the role of a sound card as far as it applies to the digitzation of cassette or LPs and ripping CD's. Is it the limiting factor for digitization of external sources.
• What are the considerations and ramifications of my starting file structure and what do I need to consider (just learned a new term Transcoding and wanted terms associated with this term).

Obviously, all these questions require research and testing on my part. What I am looking for is what kind of terms will I encounter during my reasearch. (I am not lazy, and I do know how to use a search function within a forum, FAQs and WIKI). I am looking for some basic knowledge and terms to better understand the resources you have available and help factilitate my learning. Quick solutions typically provide quickly forgetten answers, and ultimately leave you with unproductive methods for solving a new problem.

This post has been edited by Ultra57: Jun 10 2009, 10:31

[Mark7]

I think these are the important factors in choosing a codec/bitrate:
- compatibility. If you want to be able to play it on most hardware players then mp3 (NOT mp3 pro) is your best choice
- quality. Do you want (near) perfect sound or do you just want it to be understandable. If latter you can go for mono and a low bitrate

[carpman]

That's a long greedy list! So I'll try the first and last (i.e. the easy ones).

What is Sampling Rate?

http://en.wikipedia.org/wiki/Sampling_rate

What are the considerations and ramifications of my starting file structure and what do I need to consider (just learned a new term Transcoding and wanted terms associated with this term).

EDIT: WHAT IS BELOW IS PROBABLY BASED ON A MISUNDERSTANDING OF THE TERM : "file structure"

If you tag your files effectively (and name them sensibly), and you use a good software player like foobar2000, potentially you could have all the files in one directory.

If however you want additional files accompanying the audio (like images and text files) and each book spans a number of audio files then one folder per audiobook would seem sensible, something like: ..\audiobooks\author - title\*.mp3.

If each audiobook is just one audio file and you don't need additional files and you don't want to put all your files in one large directory, then:
..\audiobooks\author\title_x.mp3 , title_y.mp3 etc.

That said, I wouldn't worry too much about it, tagging the files is the key, once properly tagged, foobar2000 for example can name the files and directories automatically from the tags.

Well that's a start.

C.

This post has been edited by carpman: Jun 10 2009, 10:46

[ZinCh]

1. (see wikipedia)
2. kbps mean how large will be file, but this is not the same argument as quality
3. (I didnt use eac)
4. (I didnt use eac)
5. mhz is "resolution" of audio, it is better if it will be the same for input and output
6. it is better to leave default settings and good encoder program such as lame will do a good job
7. mp3pro cannot be played on most of portable devices, only on computer
8. 'decoder' and 'encoder' - is 2 categories of programs, 'lame' is program
9. (I dont know about nero, but) lame can produce proper results with defaults values
10. for ripping cds - it doesnt matter, foor ripping tapes/vinyl - more important is deck/turntable
11. (I dont understand)

[rick.hughes]

Normally I wouldn't comment on spelling, but in this case I think it might be helpful. The word is codec, short for coder/decoder. The plural form is codecs.

I listen to mp3 audiobooks and podcasts at least as much as I listen to digital music, so I might have some relevant experience.

You don't mention whether you will be using a portable player. Some portables seem to have a problem accurately seeking (fast forward and rewind) with unusual bit rates and sample rates. For spoken word content, seeking functionality is much more important than audio quality.

48k mono is used for many of the podcasts I subscribe to, but sometimes audiobooks will have music and even stereo content. If you are not too concerned with storage space, 64k or 96k stereo is probably a good compromise. LAME v8 or v9 should work fine for this purpose, or you can try CBR or ABR for these low bit rates if you want.

If you do intend to use a portable player, you should definitely test a few files on it before you encode your 6TB library.

If you don't have a portable but think you might get one in the future, some are better for audiobooks than others. Make sure the portable you buy will let you pick up where you left off in the book. Some portables only let you start at the beginning of a "song".

[Ultra57]

Since this is a free database accessed by over 5,000 seniors in various senior centers around the county of Los Angeles, maximization for playback versatility is a strong consideration. FYI, I am dealing with noncopyrighted material from multiple sources (i.e., LPs, 78's, cassettes, streaming media, already digitized files and some CDs). So the real question and goal is "HOW LOW CAN I GO." Part of what I do is distribute a free throw away flash drive players with a few books or old time radio programs so a senior can listen at home and then share it with another senior or bring it back to get new material. I have four computers in four senior housing centers with programs that allow them to make a CD to take home. Each of these computers has around 500 GB (and growing) of audio material. I am constantly getting new files from various sources.

I think these are the important factors in choosing a codec/bitrate:
- compatibility. If you want to be able to play it on most hardware players then mp3 (NOT mp3 pro) is your best choice

I have four PCs for this project, various stand alone MP3 CD players and quite a number of hand held MP3 players (iPOD, Zune, lots of flash drive players, and carry along style MP3 CD players) that I test for playback purposes. The codec may have been one of the culprits creating playback issues I was having with one of the stand alone players (RCA stand alone players are the worst for playback issues on MP3s). Also, as I read through the forum, I am finding that sampling rate and/or bit rate (I am still trying to determine the correct terms here) may come into play as a consideration for optimizing maximum number of playback methods. The standard among audiobook manufactuers is between 32 and 64 kbps (most downloadable books come in at 32 kpps for WMA and 48 for MP3). After reading through the forum, it appears that the MHrtz setting (sorry I don't know the term) can cause issues also and is a determining factor to consider for playback maximization.

From your answer, it appears I can eliminate all research on encoders/decoders since LAME has the most versatilily for playback. Thank you for saving me time in this area of research. I was researching under the assumption that all MP3s were alike in playability.

- quality. Do you want (near) perfect sound or do you just want it to be understandable. If latter you can go for mono and a low bitrate

If this 6 TB database of audiobooks was just for me (and even if it was not), the important factor for audiobooks is clarity of the spoken word along with the ability to at least hear the performed accent. I have found that a multitude of problems (tinny original or faded sections) can be solved with speakers or a volume change in the file (normallization does not appear to accurately fix a fading issue as well as a volume change will fix it). I believe that the manufacturers have hit on a good range (maximize at 96 kbps), and will concentrate my research and testing efforts in this range. Issues of variable bit rate vs. constant bit rate vs. average bit rate may create a playback issue also. Considering that most manufacturers choose a stereo constant bit rate setting, I will do more testing and research in this area. They must know something about playback issues and I have enough players to test for potential problems. Variable (to cut the file size down) and mono (not sure what effect this has on the end file) may be great for some players, but since I need to maximize my playback options, I may not have these options available to me.

[htp]http://en.wikipedia.org/wiki/Sampling_rate
[/htp]

Thank you for the link. I am currently reading about EAC and it’s set up. I have copied the WIKI page and printed it out for my next reading project.

NAMING CONVENTION - As far as naming conventions for my files, I had to be organized about my files to get to 6 TB and know what I have. I spent many years in publications and developed a plan of action, but we can always learn. FYI my standard naming convention for audio book folders is Last, First - Title - Reader Year. With this naming convention I can copy the directory with a utility called Path and paste it into a spreadsheet that helps organize all the files I have. Folders are organized by genre on the drives. I may need to change this since the collection has grown to big for the comfort of windows. Radio is handled by genre, program title, and then publication year (i.e., Detective/Dragnet/1957, Western/Gunsmoke/1958, etc.). I then can print out a listing of the titles available.

TAGGING/RETAGGING - I use a great program that was designed specifically for audiobooks called MP3 Book Helper that is very effective and lightweight. I am still researching issues that could arise over V1 vs. V2 tagging in regards to playback (will those cheap players just ignore the extra information or will they lose functionality with the extra information).

Since length of file name is an issue with many players, I have a very short file name for all the MP3 files in the collection. The key organizing factor for this large collection is found in the folder naming convention I have set up on the various drives.

5. mhz is "resolution" of audio, it is better if it will be the same for input and output.

How would I determine that they match (properties will tell me most of this information). And, what is a reasonable range for deviation between the two. How do changes made to resolution effect the final file. What should I expect if let’s say I need to change the resolution to increase playback options.

Normally …

If you do not tell me, then my searches will be fruitless and frustrating. Thank you.

You don't mention whether you will be using a portable player. Some portables seem to have a problem accurately seeking (fast forward and rewind) with unusual bit rates and sample rates. For spoken word content, seeking functionality and ability to return to where you left off is much more important than audio quality. … Some portables only let you start at the beginning of a "song".

Extremely important issues that I have thoroughly researched and solved. My solution was to track them in 15 to 30 minute segments. The nifty “start where you left off” function is a bit short sighted in my opinion considering that when you want to backtrack to relisten to a missed clue in a mystery, fell asleep during playback, or some other unforeseen interruption that prohibited the user from stopping playback. To backtrack more than 30 minutes is just a nuisance and obviously set up by someone that is sitting in a cubicle not really thinking about the ramifications of the idiotic decision to make track breaks over 30 minutes long. Obviously with 6 TB of audio, chapterization is just a pipe-dream and too labor intensive with the software options available at this time.

This post has been edited by Ultra57: Jun 10 2009, 21:47

[EricPost]

I just ripped a lot of audio books and this is what I did

I used EAC secure rip to WAV.

Then I used iTunes to convert it to m4a and used the 64 bit rate. You can go between a 32 and 64 bit rate and get good results, depending on your ear.

I have an iPod so it's just easier to use m4a format which I can easily change to m4b. An iPod will play m4a but with m4b extension it will bookmark and automatically place the audio book correctly in the iPod.

For mp3 I would use LAME codec and use EAC to rip. It's really easy to set EAC to rip and convert automatically to mp3 at the bit rate you want.

I use mp3Tag to modify any tags on the mp3s. That is a free program.

For my audio tapes and other types what I did was go to monoprice.com and got a audio cable very cheap. Then I used another free program called Audacity and used the headphone jack on my tape player to my computer "line in" and I got great results.

The key to recording with tapes to computer I found was to look at Audacity and in the upper RIGHT corner it gives you a level. And you have to make sure that bouncing red line never goes above the max. Also make sure when you record with Audacity you set the recordings to mono, otherwise it sounds awful.

You might also want to play with dbPoweramp. It has free features and pay features but the free ones are often just easier to use for conversion of a lot of files.

[DVDdoug]

I'll take a shot at a few of these questions...

I have over 6 TB of audiobook files needing conversion

Since we don't know the format, the size of your original files doesn't tell us much... Since you mentined EAC, I assume these are regular audio CDs? In that case, the MP3s can probably be 1/10th the size.

What is Sampling Rate?...

What does the MHtz Sampling Frequency refer to and what are it's ramification on sound quality in the area of speech?

With lossy compression (i.e. MP3), there is usually no need to worry about sample rate. The MP3 encoder will keep the same sample rate as the uncompressed file.

Sample rate is the number of sample-points in one second of sound. 44.1kHz means there are 44,100 samples per second. This is per-channel, so a CD (always stereo) has a total of 88,200 samples per second.

With uncompressed audio, we usually don't worry about the bitrate (kbps), but from the above information you could calculate the bitrate... We know CDs are 16-bits, we know CDs are stereo, and we know there are 8 bits per byte...)

For digital audio, you need at least one sample-point for the top-half of the "wave" and one sample-point for the bottom-half of the wave. So, in order to reproduce a 20kHz sound wave, you need a sample rate of 40kHz or more. (20kHz is considered the highest frequency humans can hear.) The CD sample rate is 44.1kHz. If I recall correctly, digital telephone uses a sample rate of 8kHz, which means you can't transmit any signals higher than 4kHz, and that's considered "good enough for voice".

With uncompressed audio, a higher sample rate means more samples, and therefore bigger files.

There is an introduction to digital sampling on the Audacity website.

What does that kbps rate mean to the outcome to my audiobook files and what a best starting point?

Higher bitrate = bigger files = higher quality.

The bitrate tells you how many bits are used to store one second of sound. With lossy compression, more data is thrown-away at low bitrates.

You can estimate file size with the following formula:
File Size in MB = (Bitrate in kbps x Playing Time in minutes) / 140

Since I don't work with spoken word files, I can't give you a starting point. But, spoken word is less critical than music, and you can get-away with lower bitrates.

Other than a critical ear, what evaluation programs are available to test resulting MP3 wave files and how do I interpret the test results?

Since lossy audio compression relies on psychoacoustics (human perception), listening is the only way to evaluate quality.

Reference is often made to a V0 or VX setting for the LAME encoder in the forum. What does this mean and is it part of the now mysterious set up of EAC?

These are LAME settings (and EAC can use the LAME encoder). Its a way of letting LAME set the bitrate for you... You set the quality-level, and LAME chooses the right bitrate. LAME will choose a higher bitrate with music and a lower bitrate with voice. (And, it's a variable bitrate so complex parts of the file will use higher bitrates, and silent parts will be encoded at the lowest bitrate, etc.)

Decoder vs. Encoder vs. Other - I have assumed (which means I am probably wrong) the MP3 Decoder is used when you start with an MP3 file and a MPE Encoder is when you want to make an MP3 file. Other is just that right now. What other things are used/needed to create an MP3? Evidently, LAME is an other.

Encode = compress. An encoder is used to make an MP3 (or other compressed format). LAME is an MP3 encoder, it's not a decoder. (It's not a true CODEC.)

Decode = decompress. When you play a compressed file, it first has to be decoded. Windows comes with an MP3 decoder. (Actually, It might come with Windows Media Player.)

The encoding is the tricky part... The encoder has to use psychoacoustic modeling to figure-out what data to throw-away. Some encoders are better than others. The decoder has an easier job. It only has to reconstruct the audio from whatever data is fed into it. All MP3 decoders should sound alike.

I have attempted to read the Nero manual with all it's technical information on presets. I was curious about some of the terms that are obviously tied to old audio editing techniques (i.e., volume control vs. normalization, anti-ticking vs. anti-clicking vs. anti-hissing and their role in maybe eliminating the clicks from a scratched CD, etc.). Where can I read up on this type of information to make more informed decisions on the best tool and settings for these correction techniques.

That can get complicated. It's probably best to take those problems one at a time as they come up.

What is the role of a sound card as far as it applies to the digitzation of cassette or LPs and ripping CD's. Is it the limiting factor for digitization of external sources.

The soundcard has an analog-to-digital converter that's used for recording LPs and cassettes. Some soundcards can be noisy, and noise is the usually the biggest quality issue. You need to set your levels correctly... high enough for a good signal-to-noise ratio, but not too high, or you'll get clipping (distorted flat-topped waves).

CD ripping is fully digital, and does not use the soundcard.

This post has been edited by DVDdoug: Jun 10 2009, 22:23

[rick.hughes]

“start where you left off” function is a bit short sighted in my opinion considering that when you want to backtrack to relisten to a missed clue in a mystery, fell asleep during playback...

I sometimes fall asleep. In addition to seeking and "resume" in a track, some players have "bookmarks" that allow you to save one or more starting places.

Breaking a book into smaller pieces is also something I do.

[Ultra57]

I used EAC secure rip to WAV.

Why not a direct rip to MP3? I am still reading about EAC and the "secure rip" is around page 100.

Then I used iTunes to convert it to m4a and used the 64 bit rate. You can go between a 32 and 64 bit rate and get good results, depending on your ear.

I do deal with quite a few users that have an "high-end" iPOD. Conversion to their audiobook format is relatively easy and often useful. Thank you for the heads up.

I use mp3Tag to modify any tags on the mp3s. That is a free program.

If you deal with audiobooks, you might want to look into another free program called MP3 Book Helper. They have a very slow site, but worth the wait. They also have a monstrous level of documentation (printout of the HTML pages was around 100 pages), but in my opinion, this is a good thing. Very lightweight. Drawback is that it only runs in Windows XP. However, it appears promising with the Windows 7 Virtual XP.

For my audio tapes and other types what I did was go to monoprice.com and got a audio cable very cheap. Then I used another free program called Audacity and used the headphone jack on my tape player to my computer "line in" and I got great results.

You might consider one of these jacks that has a "ground loop isolator" on it. I found my four at Radio Shack after blowing up a sound card due to a power surge coming from the audio equipment. I don't know if this helped because a ground loop isolator is designed to eliminate hissing from electrical something or other, but I feel safer.

The key to recording with tapes to computer I found was to look at Audacity and in the upper RIGHT corner it gives you a level. And you have to make sure that bouncing red line never goes above the max. Also make sure when you record with Audacity you set the recordings to mono, otherwise it sounds awful.

I know this is the standard "just do it" instructions I have read throughout the net. BUT, I have found that a stereo recording will allow the capture of both left and right sides of the recording. From there, I can cut and paste left to right or right to left of the bad track to create a clean "stereo" file. In other words I can capture the best side available on the tape. With mono settings, I am not able to accomplish this. I hope that made sense. This also may be what audacity is doing automatically. There are also tapes that have Left as part one of the book and Right as part two of the book (In the seventies some manufacturers of produced books that playback instructions were to "Play on the left for part one, then Play on the right for part two). With stereo recording, I am able to digitize this type of tape in half the time.

Speaking of half the time, Question - is there a way to feed the tape in at say "high speed dubbing" speed, and then adjust the file on your computer for normal playback? So you are fixing a MP3 file that sounds like mickey mouse on speed to sound like a real reader. Note the lack of any technical terms here, why a google search is fruitless for this type of project.

You might also want to play with dbPoweramp. It has free features and pay features but the free ones are often just easier to use for conversion of a lot of files.

What free features of dbPoweramp do you feel like recommending? Long ago I purchased a program called Switch that uses the LAME program for converting numerous file types to MP3 (Real Media file conversions were the primary reason for purchasing this program).

EricPost, Thank you for your time and knowledge. I hope my experiences and suggestions will help you in your endeavors. I do a lot of conversions from some incredible sources due to my interaction with seniors and a deceased husband that worked for a radio station.

[quote name='Rick.Hughes' I sometimes fall asleep. ]We all do. That's why it is insane to think that audiobooks should be marketed as they are today. Especially for the iPOD with the "back to the last track" nonsense limitation on their players. Annoying design flaw in my opinion. I am considering getting a Netbook (I believe that is the right term). They are those small, handheld PCs designed to get your email and maybe listen to some BBC radio. The key thing that made me want to look into them is the abilitiy to listen to BBC radio. This means that files could be RM or RA files that are VERY lightweight and perfect for audiobooks. That's my next research project.

This post has been edited by Ultra57: Jun 10 2009, 22:52

[krabapple]

dbPoweramp, like EAC, allows 'secure' ripping and 'direct' conversion to mp3 (or other formats). It's got a more user-friendly interface than EAC. You don't need to manually set up conversion parameters in dbPoweramp on a 'command line' type interface, where you may have to in EAC.

Both tools can use the Accuraterip database to make 'secure' ripping even faster. dbPoweramp automatically checks to see if your loaded disc is in Accuraterip; for EAC (at least the version I used) Accurate rip was an add-on function.
There are newer versions out than the one I have, so maybe now Accuraterip checking is automatic in EAC by default. With both pieces of software, uyou have to feed it a few discs that are already in the Accuraterip database, in order to 'calibrate' your setup.

Accuraterip is basically an online database of 'numerical' (CRC) values for other people's rips. The way accuraterip works is, if your rip's CRC matches the CRC for for the same track in Accuraterip, the rip is considered secure (because the likelihood of getting a match by chance is low). Every so often you upload your own rip logs to Accuraterip in order to contribute to its statistics, thereby making it even more reliable.

I don't know if your audiobooks are in the Accuraterip, so I don't know if that will save you time.

This post has been edited by krabapple: Jun 10 2009, 23:59

[Ultra57]

I'll take a shot at a few of these questions…

Thank you for your time. A very useful post in many ways. You have answered questions and included clear identification of terms. You have answered questions and included clear identification of terms to assist in my research efforts.

I have over 6 TB of audiobook files needing conversion

Since we don't know the format, the size of your original files doesn't tell us much... Since you mentined EAC, I assume these are regular audio CDs? In that case, the MP3s can probably be 1/10th the size.

I will try this as a rule of thumb for CD rips.

What is Sampling Rate?…

I have collected many of the old time radio from DVD distributions, streaming media, and downloads put up by other fans of the genre. As far as audiobooks, some come from downloadable sites (Audible, Simply Audio, etc.), and some are shared from other sources that I no longer have access to the original CDs. Most are in MP3 format and may require a transforming from MP3 with whatever encoder to MP3 using LAME to increase playability options.

What does the MHtz Sampling Frequency refer to and what are it's ramification on sound quality in the area of speech?

With lossy compression (i.e. MP3), there is usually no need to worry about sample rate. The MP3 encoder will keep the same sample rate as the uncompressed file.

Thank you, this does speed up my research efforts, and explains a little bit of the terminology.

Sample rate is the number of sample-points in one second of sound. 44.1kHz means there are 44,100 samples per second. This is per-channel, so a CD (always stereo) has a total of 88,200 samples per second.

With uncompressed audio, we usually don't worry about the bitrate (kbps), but from the above information you could calculate the bitrate... We know CDs are 16-bits, we know CDs are stereo, and we know there are 8 bits per byte...)

For digital audio, you need at least one sample-point for the top-half of the "wave" and one sample-point for the bottom-half of the wave. So, in order to reproduce a 20kHz sound wave, you need a sample rate of 40kHz or more. (20kHz is considered the highest frequency humans can hear.) The CD sample rate is 44.1kHz. If I recall correctly, digital telephone uses a sample rate of 8kHz, which means you can't transmit any signals higher than 4kHz, and that's considered "good enough for voice".

With uncompressed audio, a higher sample rate means more samples, and therefore bigger files.

There is an introduction to digital sampling on the Audacity website.

I have copied the link to my bookmarks and will get it into a printable form for further reading. Thank you for the links. I am curious as to the functions of this program and this will allow me to research the possibilities.

• You have included information for a good rule of thumb for ripping CDs (1/10th the size).
• One term you used that I am curious about is psychoacoustic modeling that I plan on adding to the research list going on the assumption that this is the underlying logic behind the basic algorythm.
• Regarding the sound card, I assumed it was the case that the "CD ripping process was fully digital" but was not absolutely sure. With the terminology you use, I doubt I will forget it quickly.
• Nero Question Still Looking for Direction THE COMPLAINING PART - Keep in mind that I am attempting to actually understand the Nero manual in order to use it effectively (develop an understanding of the different "tools" for fixing audio along with ramifications of the various settings). The terms are obviously geared towards someone who has manually attempted to fix files and is familiar with the various tools that came out over the initial time period during development of these tools. The manual assumes leaps and bounds about it's reader and leaves the uninitiated still wondering what this program does (At home I am going through a big tirade on what a piece of trash this undocumented, unsupported and thank you very much Microsoft for setting such a mediocre standard for other software companies to live up to. OK, I am back now.).
• Nero Question - THE ENOUGH COMPLAINING PART - Let’s start with - I have played with "Normalization" to fix where a reader sounds like he is periodically stepping away from the mike (I called this fading which could mean something totally different in the audio industry). In my mind Normalization's description in the Nero manual sounds like the PERFECT solution. When in reality my ear can not hear that Normalize does squat to fix this type of problem. The Volume tool is much more effective in stretching those peaks to an audible level. This is labor intensive though and requires a section by section fix of the file. It would be nice, and I cannot believe it has not been automated yet, to highlight the whole file and perform some function that would raise the sound level on the bad sections and leave the good sections intact. []bIf you have a good link to a site that has some of this terminology explained in simple terms, I would appreciate the referral.[b]

Thank you again, your post was very helpful in my quest for knowledge.

[Ultra57]

dbPoweramp, like EAC, allows 'secure' ripping and 'direct' conversion to mp3 (or other formats). It's got a more user-friendly interface than EAC. You don't need to manually set up conversion parameters in dbPoweramp on a 'command line' type interface, where you may have to in EAC.

This often has the unfortunate outcome of limiting the power of the user's control over what the program does. I may not need to tweak the actions of dbPoweramp and may be perfectly happy to forego control, but that will require research on my part. I thank you for the information regarding the correlation between the two programs. I would like to know if you have used dbPoweramp to convert from let's say a flac file to a lossy MP3 file? Is this part of dbPoweramps arsenal of tools available to the user?

Both tools can use the Accuraterip database ...

Accuraterip was a one week research project entailing issues such as what does it do, how does it work, is the cluck worth the buck ... Unfortunately, as you guessed further in your post, audiobooks are rarely found in their database which makes the add on unnecessary and a waste of resources when ripping audiobooks from CD.

That's a long greedy list!

Well the list has whittled down to four items in a short period of time. What a great group of friendly, helpful members.

Thank You Thank You Thank You Very Much

Thank you all for your help in helping me learn. I now feel confident that LAME is the way to go for MP3 conversions due my functionality lmitations. With the help from the kind members here, I have limited my research time greatly. My research is still ongoing in regards to

• EAC, Still Reading the Onlilne WIKI and FAQs in printed form
• Transcoding Next on list for researching types of and considerations when transcoding from different file types (Going from MP3 to MP3 is a major concern since I have many in this format that may be unusable in their current condition)
• Find a way to learn those audio industry terms used in Nero's now cryptic manual.
• Futher reading, but primarily testing of the kbps within ranges of 48 to 96 kbps (keeping in mind the suggested >32) .

[rick.hughes]

[*]Transcoding Next on list for researching types of and considerations when transcoding from different file types (Going from MP3 to MP3 is a major concern since I have many in this format that may be unusable in their current condition)

I have sometimes needed to transcode MP3 to MP3 due to the seeking problems I mentioned before. With speech the loss of audio quality (sometimes imperceptible anyway) from transcoding lossy to lossy is not nearly as important as it is for music.

I use dbPoweramp. It will let you change the sample rate and bit rate if you need to, go to/from stereo/mono, to/from different codecs such as FLAC/MP3, apply ReplayGain, and lots of other features. Of course, many other applications will perform these functions as well, since mostly the codecs are doing the work and the other applications are just user interfaces to the codecs.

You mentioned needing to fix the volume in some cases. I don't know of anything that can automatically and intelligently decide when some part of a track needs to be made louder, but MP3Gain and ReplayGain are some things that you might look into if you haven't already. They will allow volume changes to tracks or "albums". This might be what Nero Normalization is doing. The volume of the whole track or album will be adjusted up or down by a certain amount.

Some more comments on portable players. I have found the ability to save multiple bookmarks very useful. I can switch from podcast to music playlist to one or more audiobooks and always pick up where I left off. The player I currently use is the Trekstor Vibez. It is "descended from" the Rio players such as the Karma that also had this feature. Unfortunately, neither the Vibez or any Rio players are available anymore, but some may be available used on eBay for example.

Another player I have researched that seems to allow multiple bookmarks is the Cowon D2. but I don't have any personal experience with it.

Finally, if you are not already overwhelmed by information, there is Rockbox. This is alternative "firmware" for portable players that can often add additional functionality.

[Ultra57]

I will look into dbPoweramp and thank you for the link. I appreciate the information you have shared about your personal experiences with transcoding from MP3 to MP3. I am sure they will be most helpful in my project.

mp3Gain and replayGain
If I am understanding your post correctly, the "fading" in and out issues I am trying to deal with are unfortunately on portions of a track (not the whole track) and can not be solved with the suggested programs (this is a question)? It would work if it makes the bad sections of a track louder and leave the other sections alone, but from your description, I appear to be stuck fixing these files by hand with a wave editor program (labor intensive). The description of normalization appears (at first glance) to address this kind of problem, but the results are simply annoying (it stupidly normalizes the hisses and ticks associated with cassetes by amplifying them to a more audible level). There may be a setting that I am missing within the interface, but without a proper understanding of the terms, it is a hit and miss situation with strictly missing going on and no hits. Other people I have talked to seem to agree that we are all stuck fixing things the old fashioned way ... adjusting the volume in real time with the amplifier/tape player controls or editing with a wave editor. It is hard to believe that this procedure could not be automated ... hunt down any low level sections (based on a time based parameter set by the user) and amplify the volume level for the found section(s).

I use VBR - MONO Settings
I cannot tell you how many times I have read users rant on about the stupidity of a user not using these two settings for MP3 audiobooks. You were much more polite in your presentation of this suggestion. I am aware that VBR and Mono are typically touted as the most "correct" or the best settings for audiobooks, but it would be naive of me to assume that publishers are stupidly using bad settings that could easily be changed. Keeping in mind that since publishers are distributing these books at between 32kbps to 64kkbps for WPM* and at 48kbps for mp3 with CBR and Stereo settings, I must seriously consider the ramifications of changing these settings. Publishers must know something because their livelihood depends on producing the most usable product in order to capture the market and stay in business. I must also take into consideration that this change could potentially decrease their bandwidth costs, leaving me once again concerned that there must be a reason they have not changed their settings to the more "preferred" settings of VBR - MONO. I will need to do some simple tests to determine the ramifications of deviating from those settings.

*Reportedly, WPM performs better at lower rates than MP3 for audiobooks. Who knows, maybe it is the same tech making the bad track breaks over there at Audible (just kidding).

[rick.hughes]

... 48kbps for mp3 with CBR and Stereo settings...

This may very well be the best general purpose settings for spoken word content because:

* 48k gives the smallest file size likely to be compatible with the most players including very old ones.

* CBR is likely to be compatible with the most players including very old ones.

* Some content really does have stereo, with different "actors" speaking from different places.

[DVDdoug]

You have included information for a good rule of thumb for ripping CDs (1/10th the size).

That was really just a ballpark guess on my part, and it's only for voice files (and I don't have a lot of experience with spoken word files). Good quality music MP3s are going to be in the ballpark of 1/5th of the uncompressed size. From the bitrate suggestions posted by others, you'll probably get files less than 1/10 of the original size.

One term you used that I am curious about is psychoacoustic modeling that I plan on adding to the research list going on the assumption that this is the underlying logic behind the basic algorythm.

Right... But, you really don't need to know the details. A lot of it has to do with "masking", which is when one sound drowns-out another. Like when you're at a rock concert, and you can't hear the person next to you speaking. The sound waves are getting to your ears, but your brain can't detect the speaking over the music. Here is an article about how the MP3 CODEC works.

Nero Question Still Looking for Direction THE COMPLAINING PART - Keep in mind that I am attempting to actually understand the Nero manual in order to use it effectively (develop an understanding of the different "tools" for fixing audio along with ramifications of the various settings). The terms are obviously geared towards someone who has manually attempted to fix files and is familiar with the various tools that came out over the initial time period during development of these tools. The manual assumes leaps and bounds about it's reader and leaves the uninitiated still wondering what this program does (At home I am going through a big tirade on what a piece of trash this...

I don't use Nero... Nero does a million different things, and you're probably right about the "user manual". But, it's a tool... You can buy a toolbox full of Craftsman tools, but that doesn't make you a mechanic. And, even if you know the basics of how to use the tools, it doesn't mean that you can diagnose and repair your car. Audio is a special field. Like any specialty field, it has it's own terminology/jargon and a body of associated knowledge.

The basics of audio editing are cutting, splicing, fading, crossfading, etc. Then there are effects like mixing (blending two ro more files/signals), equalization (fancy-precise "tone control"), reverb, etc.

Things get more complicated if there's something wrong. The sound might be muffled, or you might have noise. Different kinds of noise need differerent filtering/repair techniques. And, almost all of this stuff requires trial-and-error. Sometimes "the cure can be worse than the disease", and you're better-off doing nothing.

When you run across a problem, make a new post (this thread is getting rather "involved" already), and try to describe the problem. I know this can be difficult when you're new to the terminology... It can be hard to describe some audio porbelms when you do understand the terminology! And, tell us what the source format is. The kinds of noise you get from a vinyl record is different from the kinds of noise you get from a cassette, etc.

Nero Question - THE ENOUGH COMPLAINING PART - Let’s start with - I have played with "Normalization" to fix where a reader sounds like he is periodically stepping away from the mike... Normalization's description in the Nero manual sounds like the PERFECT solution. When in reality my ear can not hear that Normalize does squat to fix this type of problem.

I don't know what the Nero manual says, but "normalization" can have two different meanings... The usual definition means that the volume is digitally maximized, but the whole file is adjusted by the same amount. With music, this is what you want... You want to keep the quiet parts quieter than the loud parts. I assume tha't what Nero is doing. But sometimes, "normalization" is an "automatic volume control" that turns-up the quiet parts, and turns-down the loud parts.

There is a FREE program called The Levelator, which might be exactly what you are looking for! (I wouldn't recommend running all of your files through The Levelator, but only where there seems to be a defect. Sometimes the reader may intentially modulate his voice for expression, and you might not want to wipe out that effect.)

[Ultra57]

... 48kbps for mp3 with CBR and Stereo settings...

This may very well be the best general purpose settings for spoken word content because:

* 48k gives the smallest file size likely to be compatible with the most players including very old ones.

* CBR is likely to be compatible with the most players including very old ones.

* Some content really does have stereo, with different "actors" speaking from different places.

Research on best settings may be hampered because this isn't a medium that is being played for an audience on a pair of Altec theatre speakers using a Karmen Harden amp ... for a giant block party. I can just see the flyers "Rock Out with the Newly Remastered Archangel's Hamlet - This Saturday at 5 pm - BYOC (bring your own chair)."

Also, lazy online researchers with small or insignifigant collections (as determined by a particular forum group) may be getting a quick fix from serious resources and touting it as the "only way." Thank goodness we are not mindless computers crunching out our audiobooks and loosing the stereo effects or other aspects of the original.

Thank you for sticking with me.

[Ultra57]

The links provided direction to some very nice printed resources. That along with an Amazon book search has provided me a broad base of good choices. This is also probably the most fruitful direction for my research to continue.

Since experience is the best teacher, it may be a while before I return. Because you have all have been kind enough to give me a good starting place, I'll try to stop by with results or observations on the various aspects of my findings along with ABX testing results and samples.