ACXU Presents: How to Pass ACX QA Every Time

[Music] welcome to ACX University I'm Scott Jacoby and today's episode is how to pass QA every time we'll be joined by Brendon and David from the QA team and they'll review the top reasons your titles might get flagged during our review process let's get started hi my name is David Afonso I'm an audio engineer with the ACX Quality Assurance team and I've been working with ACX for two years now prior to joining ACX I worked as an audio editor and recording engineer over at the performing arts organization Symphony space and Galaxy Smith studios located in Brooklyn New York I'm here today to discuss the importance of editing and encoding your audio files and why these two requirements should never be overlooked we'll go over some of the common issues and challenges ACX producers face within these two stages of the production process I'll cover what causes these issues and offer some tips on what you can do to address them so AC X editing requirements are there to ensure that overall sound quality meets AC X and audible standards an overall lack of quality control will lead you to run into some common issues that directly affect how well your book will be received by your listeners some of these issues include narration with extra mouth noise and vocal artifacts long gaps of audio silence within the middle of a file clicks and pops at the beginning of a file and at the end of the file and this happens more than you may think and lasts a loud noisy recording environment all these issues can really distract the listener from not only the story but your performance as well so first cause being a poor recording environment the acoustic space in which you record in really plays a crucial role in capturing a good quality vocal sound we must keep in mind that microphones are highly sensitive therefore they're not only going to pick up the sound source in which case being your voice but also reflections from solid surfaces and noise from the outside some examples of these unwanted sounds can be reflections from the walls surrounding you from the ceiling from the floor outdoor traffic noise doors being opened and shut or maybe just the dog barking next door in the apartment all these unwanted sounds can really make for a poor recording environment and should be considered before any production begins the second cause would be gear selection specifically in regards to microphone choice the recording environment in which you record in should dictate what type of microphone you should be using selecting the right microphone for your specific environment will have a major impact on minimizing unwanted background noise caused number three would be poor microphone placement and technique a performance can be very dynamic in nature from soft passages to loud passages positioning yourself in close proximity to your microphone and speaking too closely to the capsule will result in plosives accentuate sibilants and cause distortion caused number four is not utilizing a quality control sheet a lack of familiarity with our requirements and not knowing what to look out for will result in your files containing miss status such as repeated lines missing content loud breaths etc having a QC sheet on hand will help keep track of these issues and later on you could just go ahead and fix them in post-production so the first preventative measure that you can take is to record in a non reverberant quiet room getting a clean vocal recording can be challenging when trying to select a recording environment in a domestic household so what you want to look out for are soft furnishings and carpets my suggestion is either your living room or your bedroom the second preventative measure that you can take is to make sure that there's enough distance between the sound source your voice and the microphone this will prevent pops loud breaths and unwanted vocal artifacts from entering your recording a good place to start is the hang ten sign this is a great way to measure your distance on the fly so that provides safe distance between the pop filter your microphone and your your voice in addition to this you can vary your distance from the microphone and this is to accommodate a dynamic performance this helps keep levels even through soft passages and loud passages third preventative measure that you can take is using a dynamic microphone as opposed to a condenser microphone especially if you're recording in a noisy environment dynamic microphones are more durable are going to offer better protection towards plosives and carry a frequency response that is more suitable for noisy environments this is why it's the ideal choice in any live application some popular mic choices that fall into this category are the electro voice re 20 and the Shore sm7b for more information on these types of microphones I suggest checking out the ACX youtube channel we have a great video there on gear selection preventative measure number four is using a pop filter placed in front of the microphone this will help in taming plosives and sibilants this is a tool that is a must-have for any audio professional looking to achieve a clean vocal sound and is the best way to address these issues at the source v preventive measure that you can take is punch and roll recording this can help as recording through an entire chapter will most likely result in missed edits repeated lines outtakes and loud breaths for example if a phrase is giving you trouble you can set up a recording loop to start a few lines before it begins to record at the trouble point it often helps the performer to hear the previous lines to help with consistency in their delivery and allows them to get back into character and focus more on the performance preventive measure number six is to record and save a clip of clean room tone that way you have it at your disposal to use if you need to fix any issues in post-production the last preventative measure that you can take is utilizing a QC sheet to identify and resolve any editing issues a checklist listing our requirements will guide you through the entire process of cleaning up your audio this ensures proper revision before submitting your files to ACX so that means checking for clicks pops and background noise within the spacing at the beginning and end of your files it includes removing any apparent sections of silent audio within a file and also removing any long breaths or excessive mouth noise short breaths are okay and provide a natural feel to the performance now in the case you receive a notice from a CX identifying that your files contain some editing issues how would you address these assuming you saved a clip of clean room tone you would simply locate the problem area and paste a necessary amount of clean room tone another way of dealing with long gaps of silent audio locate the area and refer to the manuscript to make sure that the file isn't missing any content once confirmed you can then simply just shorten the gap please do keep in mind although you may be tempted to use some processing tools such as EQ noise reduction or gating these should be used as a last resort as improper use of these tools can seriously damage your recording so a CX encoding specifications are required to ensure that your files are optimized for quicker uploads faster downloads and to facilitate distribution across multiple different platforms so some of the causes as to why you may be experiencing some encoding issues include if a file exceeds 120 minutes in length or 170 megabytes in size if a file is not encoded at a forty four point one kilohertz sampling rate or if a file was encoded using variable bitrate as opposed to constant bit rate constant bit rates allows for quicker processing of your audio data and allows for better prediction of file sizes another cause can be if your audio book consists of a combination of stereo and mono files so you want to be sure to deliver all your files either in mono or either all in stereo if the production consists of only narration then it is strongly suggested to submit your files in mono the human voice is a single sound source and therefore there is no need for a stereo track recording so you can prevent yourself from running into these issues by double-checking and revising these settings and refer to the a CX encoding guide before exporting your files from your recording software our encoding guide is available for download from the a CX website and guides you step by step through the entire process please check it out in addition to this most if not all recording software allows you the option to save an encoding template this is highly recommended and can be used for future productions in the case you must re-encode your audio files you always want to make sure to go back to the original source audio wav files and reprocess with the correct settings by once again referring to the ACX encoding guide reprocessing mp3 files will result in a huge loss of quality for more information and an in-depth look into encoding I suggest checking out the ACX blog for a post titled encoding audio with Andrew the audio scientist let's take a quick break for the great debate where our presenters are tackling your burning questions about audio book publishing welcome to the great debate where our panelists will argue the hottest topics of today's audiobook industry on my left is Chris Shula and on my right is Dara Rosenberg each panelist will have a chance to answer for each question and I'll keep an eye on the clock to make sure nobody takes it too far let's dive right in with topic one punch record versus straight record Chris let's get your take my take is straight record especially when narrators are just getting started it's really important to stay and what I would consider to be the performance flow so if you make a mistake you make an error you can use something like the click method to click then you know in your digital audio workstation you see a big line where you know it's a very clear mistake and later on you can go and make those edits but you but the most important thing when narrators are starting is to understand what that performance needs and to stay in performance flow that's the best way to do when you're starting straight record all right straight record from Chris Dara what are your thoughts you've got a punch record it's the way the publishers want it just start from the beginning so you learn how to do it I believe it makes time way more efficient you just go go go it doesn't interrupt the performance if you're you're in it just learn it from the beginning and you're set so you think the performance is more important I do and also you have to understand that we only have a certain amount of time right to get this all done however when you're starting out that's not the most important thing most one thing is getting the performance so don't worry about time time is is you can take as much time as you want stay there just make it good Dara can you take as much time as you want no you can cuz you have a certain amount of hours in the day and you just want to get as many hours of narration that you can get done you make little short cuts however you want to make it and you go bloom bloom cut paste whatever you got to do it works out you can record faster and you can narrate faster not a nary faster just work more efficiently I guess zoom yes I guess my thing is you know when when you're starting out the most important thing is to be good you need to be good and the only way to stay you're gonna be good is to stay in stay in so later on when you learn how to get you know people who are working underneath you and subcontracting for other stuff maybe but right now you want to be good stay in be an actor so okay we have be good for blum blum you at home can decide which you prefer this leads us to our second topic for debate self edit and post produce versus outsource Darryl what are your thoughts send it out if it cost you X amount you get X amount of money for a narration and Y to send it out get as much as you can done and don't spend the time editing it and mastering it get as much work as you can done grace you're gonna send it out no not at the beginning not at the beginning one's career why because you spent a lot of money on your on your space that you're recording and you spend a lot of money on your mic you made a big investment in your audiobook career so do you have more money to send out to subcontract people to do your post-production no and you want to learn how to do it the right way you want to learn how to do post audio production so that later on in your career you get an instinct on whether or not things are going right or wrong from those people who you are giving that post-production work to Darryl what are your thoughts on the fact that the artistic brain and the tech brain might be two different brains well listen sometimes it's good for somebody else to hear your narration you might miss things you might just always think you're awesome so you're gonna miss out on things and I don't know how to master so I send it to somebody learn how to master learn how to master okay on to topic 3 DIY closet studio vs. home vocal booth Darren go ahead do it yourself I'm just gonna say I have a studio Brix but I was in a closet for six years in Manhattan and nobody noticed there was a sound change when I went to my studio Brix however you can make your closet sound good if you have the right acoustics you can work for any publisher you want Chris it all depends on the space where you live a lot of people don't have that space to be able to create a DIY space that actually sounds good we have to have we have to make sure that the environment that we record in sounds applicable to a narration if it is not consistent if you got noises coming out of the neighborhood all over the time you can't use that space so I have a whisper room and in my neighborhood I'm that gives me enough sound isolation that I'm able to record audio books all the time Chris they barely pay me anything here I don't have enough money for a home studio what we do well in that case then you want to look at the best room in your in your apartment the best closet and looking at at insulating that as best as possible to do to take away all the bounce so Darren it sounds like I've brought Chris over to your side because I'm a winner and that's what you should do alright let's wrap things up with a lightning round these questions will be fast and furious ready let's go headphones versus speakers when editing Chris headphones Darren I don't edit like standing versus sitting in the booth Gareth sitting Chris sitting paper script versus tablet Chris tablet Dara Kathleen pronunciations Chris tomato vs. tomahto tomahto Dara schedule versus schedule schedule Dara February vs. Febuary I don't know how to say that word Chris groceries vs. groceries Cherie groceries yes I don't know what that was if you're happy and you know it Dara clap your hands Chris I know I've caught the hands too yeah both clap their hands yeah man will you go out to dinner with me tonight you're handsome yes will you go out to dinner with me tonight no oh okay I've got one date at the very least sounds good to me that's all the time we've got today for the great debate join us next time as our panelists debate the burning issues of the audio bikini hi my name is Brendan sollars and I am a QA engineer with a CX QA and I've been doing it for about four years and I've listened to thousands of your books and so during this time I think I've come across a few pointers that I think producers at any level can really benefit from and we've broken some of these down into sections that I'd like to share with you today so I hope you enjoy we require that every book that is submitted via ACX have regular spacing and the spacing should match spacing that is common for books on audible each file that you submit should have between 1/2 to 1 second of clean room tone at the top and then between 1 and 5 seconds of clean room tone at the end of the file and this is to ensure that your book just flows smoothly from chapter to chapter if there's too much space at the beginning or end of each file a listener might think that gosh there's no audio playing and maybe the batteries in my device are dead or if there's too little space at the end of a file and there's something dramatic that's just happened and you're supposed to maybe think about it for a beat before going to the next chapter and you miss that opportunity it's gonna feel weird for the listener so those things are all important as far as like the listener experience goes as a QA engineer I really like files that are spaced properly because I know that if a file is spaced correctly its complete there's no error in processing all of the manuscripts should be there because I know you followed the editing standards that we have for audiobooks if the spacing is too short there might be an encoding issue or maybe you forgot something and cut the file too short so it helps me pass your title faster some causes of spacing issues might include just the fact that when we're editing we often edit with fidelity to the man's manuscript in mind as opposed to thinking about the listening experience for the user I'm you know it's it's really easy to get bogged down in the weeds of making sure that all your punches are in the right place you know there aren't any crazy breath noises that are distracting and then you maybe forget the final step which is to properly space your file another common thing that I see is a lot of digital silence at the end of a file or files that end mid word this could possibly be due to not selecting the right region when exporting if you're exporting by region and you don't select the correct timecode in your export window you can export too much audio and then have a bunch of digital silence with no room to run or anything in it or or have too little space or have your narration cut off so that's another very common spacing issue that I encounter you can begin preventing spacing issues in your AC X production by setting up your recording correctly in the setup stage so when I set up my recordings I always record about 30 to 60 seconds of clean room tone to use during the editing process and this helps me keep editing and spacing in mind for later on in the production to make clean room tone if you don't know how get your you know recording station setup the way you would when you're recording then hit record leave the room so you don't breathe on the microphone or rustle some papers or something let it roll for about a minute or two go grab a coffee come back and then listen to your audio trim it up and then I always suggest turning up your monitor gain a bit in order to hear everything more clearly if you have any repeated sounds in your room tone or use it in editing and then repeat sections of it and it has small noises in it sometimes that'll be really distracting so make sure it's extra clean turn your monitor gain back down so you don't blow your eardrums out when you start to record and then use this to edit and space your files another issue that I we encounter like I said was digital silence at the ends of the files and this is due to encoding issues whenever I am encoding my audio I first double-check the timecode of my track where I'm encoding from and make sure that the timecode that I wish to export is actually in the window for my my encoder and if there's any issue there I make sure to select the correct region and then export after I export I always double check the files by dragging them back into my D aw and quickly checking the heads and tails and this helps not only with spacing but editing and mastering issues as well you don't have to export again of course since you've already got your files in a folder somewhere but if you follow those things during your recording process you should be pretty good to go if you've received a notice for me or another QA engineer saying that your files have too little spacing at their tails use clean room tone that you may have recorded earlier in the session to space these tales out this is always the number one way to fix this problem if you don't have clean room tone you didn't follow the steps that I described earlier you can sometimes find room tone within a file just make sure that you check it using the steps I described earlier you know turn the monitor gaming up a bit listen for any background noise make sure that the noise floor is consistent with the rest of the file and you can probably use that to space out the file if there's too much space you can easily just trim down the end of the file use a short fade at the back of the file to make sure that it transitions nicely out I do recommend that if you do find too much space at the end of a file and it's not due to an encoding issue that you check the file to make sure you've edited it correctly so that's one sign that I usually find in a file that hasn't been edited correctly so it's always good to double check what is RMS and why is it important well let's first define RMS RMS stands for a root mean squared which is it's an equation that is used to find the overall average volume of a volume of an audio file and engineers use this in recording to determine how loud a file is going to sound to human ears it's really the way audio is perceived volume wise so as a QA engineer I compare your files to the audible standard which is that every file submitted to audible have an RMS value between negative 18 negative 23 dB if your productions outside of this range say it's quieter than it should be it'll sound quieter than books available inaudible conversely if the RMS is too loud it's going to sound your file is going to sound much louder than other books available on audible not only should your files conform to the individually conform to the audible standard but your production as a whole should have a consistent level throughout if users are reaching for the volume knob on their their player they're not going to be thinking about your performance so support your performance by creating files that are even in level and easy to listen to and that's why Armas is important there are a few causes that occur at different places in the recording process for low rms issues which is what I'd like to focus on today lower RMS is by far the most common issue that I encounter mastering wise during my checks so I think it's important we address low rms here today and I find that most files that have low rms that are submitted are just recorded too low and I think this may be because you know one of the first things you learn about as an engineer is that if you peek at 0 DB you're going to distort and so producers sometimes overcompensate and then keep their their mic gain too low and I used to do the same thing and I get these noisy recordings that were hard to master later on so the more level that you have at the start the easier it's gonna be to master later on one final cause of a low rms issue would be mistakes made during mastering either not knowing about mastering at all or not knowing how to measure your rms and peak levels using its a statistics tool statistics schools can be very difficult to find within a da w they're often hidden in drop-down menus and have confusing names and you know with I didn't know about a statistic tool until pretty late in my career to be honest and you can't fix something that you can't measure so if you don't have a reliable way of actually measuring RMS within your GIW you're not gonna be able to fix your problems not understanding you know the concepts of limiting and normalizing and compression can also you know make mastering your files difficult either you don't know how to do this at all or you don't know how to even out your Peaks if you can't even out your Peaks you're not going to be able to have a consistent level another common thing that I find during my checks are files that are peaking at negative 18 or negative 23 DB and I think this is because a lot of producers confuse rms and peak level so they'll see the rms level and think that their voice has to be peaking at negative 18 or negative 23 DB when really those are rms levels that they should be aiming for in the mastering section that's that happens more often than I'd like but both both rms and peak level are pretty confusing so it's it's easy to get lost in the weeds there prevention starts at the recording setup stage when I record I always keep in mind that I don't want to record too low again like I said that's one of the most common issues people come into with lower rms it's just having too low of a recording level so when I start recording I know that I want my voice to be peaking around 10 DB on my meter so I sit down I get situated and I adjust my microphone gain until my voice is speaking around 10 DB or so and once I get my level set and I'm comfortable I don't touch it again for the remainder of my recording because I once you said it you want to keep it consistent I actually put a piece of tape right next to the knob to make sure if someone comes in and monkeys around with my stuff I can always put it back where it needs to be we have a saying set it and forget it and that's really important when you're recording with level in mind if you need to adjust your monitor level while you're recording make sure to do that and your monitor stage in your headphones not with your preamp that's going to mess up your level and it's not going to be a great experience for the listener proper mic technique is very important make sure that you're the correct distance away from the microphone we use a you know about a hang tens worth away from the face of the microphone to ensure that you're at the proper distance from the face of the microphone this ensures that enough direct sound is hitting your microphone you're getting good level and also you know proper mic technique while you're narrating is important say there's there's a very exciting portion of the narration or like some yelling that you have to do try to imply dynamics as much as possible not raising your voice and if you need to you can maybe move your head back a little bit away from the microphone in order to make sure that your level stays consistent the more consistent your level is during recording the easier it's going to be to master later avoiding lower arm issues in the mastering process begins with figuring out where you need to start in the mastering how aggressive you will need to be in a mastering and you'll do that by using a statistics plugin evaluate each file that you have for peak and RMS level and then I always recommend grouping like files together in your GIW for mastering because files that have similar qualities either peak or rms qualities can be usually be mastered together it'll make things faster and easy for you in that process once you've done this I recommend using normally normalization and compression to even out any high peaks in your files that are much higher than the average level of Peaks in your files this will ensure that your listening experience is easy on the listeners ears and as much as I don't recommend looking at audio files and making a determination of how things sound it is helpful to look at a waveform and just visually see does the file look uneven if there are big Peaks you know that you'll need to use compression or eliminating to bring those down some pointers during mastering that will help you avoid level issues don't eq after you've got good level after you've limited or compressed or normalized because EQ will either cut or boost gain within your file and affect your files level a lot of people make this mistake and it gets really confusing when you don't know where in the signal chain you're making these mistakes also keep your monitor level consistent in the mastering process if you're adjusting the monitor level of your speakers or headphones while you're mastering you might make a mistake and think that your files are louder or quieter than they actually are so again keeping consistent level is important in the mastering process as well as the recording process say you get a message from me or one of the other members of the QA team concerning low rms how do we go about fixing this well first I recommend taking a look at your QA notice because this will have the files that are affected by this issue flagged I recommend running statistics on these files to see where you are in in level and how much compression or limiting you might have to do on these files if you don't know how to find RMS or peak level ask QA we're more than happy to help you navigate your da W to find statistics programs you know as I said before you can't fix something that you can't measure so that's really key to getting a great mastering performance take a look at the files if they look uneven you'll have to use limiting or compression to bring down the louder peaks and then normalization to bring them back up to the maximum allowable level don't forget after measuring the statistics of your files and before you start mastering you should be working from your raw edited files not the mp3s that you've uploaded to a CX it's always important to work from the highest quality audio files as possible Riaan coding mp3s can cause a lot of audio issues so it's best to work with the the highest quality file that you have why is title structure and formatting your manuscript for audio important and what does QA check to make sure your file falls within these requirements well title structure requirements all deal with making sure your audio is complete that there aren't any duplicate files in your audio book that it's structured in a way that is easy to listen to in audio format print books and audio books are inherently different due they're different you know formats and so some things that work in a print book don't work in an audio book during my checks in QA I checked the opening credits to make sure that they match the cover exactly if they don't that's a big problem I make sure that the next file in the book whether that's chapter 1 or the introduction file I make sure that that file doesn't have the opening credits at the top of it as well this is a big problem that I encounter quite often you know people will repeat the credits in the second file which we don't want to do it's just repeating information I checked the section headers to make sure that they're in order and that they make sense and that every section header within the manuscript is in the audio I also check to make sure that your files don't have more than one chapter in them or section basically we're just checking to make sure you know everything has been uploaded properly that all the files are complete and that it makes sense in audiobook format no creator wants you know to download their book and hear the opening credits of their book and then the last chapter immediately after that they would ruin the book and no user wants to be on their player clicking next and expecting the next chapter in in the book but then actually getting a chapter that's five chapters down the line also following these guidelines helps me check your book and move your book forward more quickly so an organized production will help your book move faster basically some causes of entitle structure issues are just well it starts from the beginning with script prep if you're not communicating with the rights holder or ironing out any inconsistencies of the manuscript those will show up in your recorded audio also if your project file is confusing say you know a lot of producers will record all chapters on one track this makes editing more difficult and it also increases the chance of having duplicate content confusing filenames can also cause issues with title structure when you're uploading if you don't have consistent file names or file names that make sense it can often lead to files being out of order for instance you could have an acknowledgments file that's supposed to be at the beginning of the book at the end of the book it's very easy to make that mistake as a QA engineer I only have the files in front of me to make these decisions so if I can't tell where a file is supposed to be in the file order I'm going to have to reach out and it slows down the process for you getting your title live avoiding issues with title structure begins at the setup stage of the recording I always recommend that you confirm both the title and author credit with the rights holder before recording the opening credits a lot can change between the kindle version and audio version of a book so it's important to make sure that you're on the same page there if the title in the audio and author credit and audio don't match the cover it'll cause issues with QA this is also a good time to let the author or rights holder know if you're planning on using a pseudonym if they want to include your name on the cover when preparing your script don't include items like the opening credits or a title page as these don't really make sense in an audiobook format record each section or chapter in a separate track in your da W this will help keep things [Music] separated for editing it'll make it a lot easier and you can also name each track with the number in which the file should be included in the book for instance when I set up my project file I always name the first project of first track zero one opening credits second one would be zero two acknowledgments this way I know what file comes next when I'm uploading and it also helps the QA team know that your book is complete and in the right order one once I had a book that was actually a time travel book and so all the chapters were out of order luckily the producer included the number of the section in the file name so I was able to sort that out pretty quickly if you have title structure issues that have already occurred and you need to resolve them before your title can go live i recommend following up the rights holder just to confirm the opening credits closing credits and any ambiguities that might be in the manuscript I always recommend with any production that I do I recommend that you do the same checklist I guess and that would be just to check the opening credits to make sure that they match the cover making sure that the credits are not repeated in the second file I would recommend checking the file order to make sure that everything is in its proper place and then you know there's no duplicated content or combined chapters if you follow these suggestions every time you record you'll have a really easy time passing QA so good luck [Music] let's take a break from the action and check in on a CSS Andrew Gallagher and his QA corner hi I'm Andrew Gallagher I'm the cue a supervisor and production of Angeles for ACX welcome to the QA corner where we take a quick look at the little details you've got to get right before you consider your audio book production complete this corner is going to cover samples [Music] your sample should be between one to five minutes we find this is the sweet spot to listen to you should find a good piece of content to fit that 1 to 5 minutes - there also can't be any explicit or offensive material within the sample this is because anybody can listen to the samples on audible and they may not want to hear that kind of material you also cannot include things like credits introductions or music we don't allow these items because they get in the way of the content of your audiobook this is the first time that many listeners will hear a part of your audiobook so you should want to have the best and most intriguing part of your audiobook as your sample you should work with your producer to find not only the best content from your audio book but also the best performance from your audio book and you should use that as your sample you want to pull the listener in and you want them to buy your audiobook thanks for watching join us next time when we take a quick look at the little details you've got to watch out for before you consider your audio book production complete thanks guys we hope your audiobooks will pass our QA process the first time every time from here on out join us next time on ACX University [Music] you

Channel: AudibleACX

Views: 32,432

Keywords:

Id: p156u60gWdM

Length: 43min 8sec (2588 seconds)

Published: Thu Sep 21 2017