The Symphonova Stems service provides you with a completely new hybrid method for creating orchestral stems.
Our intelligent software captures the musical expression of a small group of musicians that you record and applies the musicians’ expression to the virtual parts created by our system. The result is a full set of stems with authentic musical expression. Depending on the package you choose, we deliver the stems ready for your mix in several instrument grouping and mic positions.
In short, without adding any expression to the virtual parts, you’ll be able to get a full set of stems that once mixed sound like the first audio example once you sent us a recording of a single instrument, such as in the second audio example below.
audio 1.1: Violin section processed by the system using audio 1.2 below.
audio 1.2: Single violin recorded; used for processing audio 1.1 above.
The same principle applies for all instrument families in a symphony orchestra. Our service enables you to get the stems of a full orchestra by recording only a small ensemble. The musicians can be recorded remotely or in an ensemble – see below for more details about recording methods.
We offer three packages. In all packages we deliver the stems in three mic positions: close mic, conductor’s tree, and surround. In our Full Orchestral Suite package, stems are delivered in two additional positions, which are: super-dry and hall.
All of our stems are 48kHz 32bit recordings, with levels balanced to give you a starting point for your mix. This puts you fully into the driver’s seat to blend, add sound effects or non-acoustic instruments; and especially, to use your own sound-stage, acoustics and reverb to create your own individual, signature sound. You can be sure that our stems give you all the freedom to celebrate your musical individuality, and will not force your music to sound like it was created on a production line.
Using our recommended recording methods, another unusual feature of our stems is the absence of bleed between sections; especially on our conductor’s tree, surround, and hall mic positions. You might be producing the last pages of Mussorgsky’s Great Gate of Kiev in the Ravel’s orchestration, but even the surround and conductor’s- tree stems of our woodwind and string sections will have only the faintest hint of brass. Our stems therefore enable you to highlight instruments in your mix as if you had recorded your instrument sections on their own.
Our pricing is based on only two considerations: the length of your score, and your choice between our three packages of stems. Full details of our pricing and stem-packages can be found here.
The Symphonova Stem Service is a very powerful tool that will respond with high sensitivity to whatever you give it to work on. I strongly recommend that you take the time to go through the tutorials in a systematic way so that you fully understand how to use our service and how to get the best results from it.
To summarise, our service transforms:
I will now briefly cover our three core concepts:
Depending on your score, you can get high quality stems by recording from 1 to 12 musicians. This works on the basis of recording appropriate parts of your score, which we call the Expression-Leaders. These recorded parts lead the expression of the virtual instruments. In the process of score preparation, you select the parts that are expression-leaders for the virtual parts that are created by our system (the expression-followers).
Each instrument you record can provide you with the family of that instrument. For example, in a score that includes two oboes and an English horn, the principal oboe part would be recorded, and on the basis of that single part, the resultant stems would include the second oboe and Cor Anglais. So in that example, the first oboe is the Expression-Leader, and the second oboe and the English Horn are expression-followers.
Each part recorded by a musician can also be selected to lead the secondary instruments of any other instrumental family. (This is accomplished in your MIDI score using CC17. The complete list of CC17 values used to select Leaders can be downloaded here, and a full tutorial can be found here).
This means that, on the basis of only two recorded instruments from each full section (strings, woodwinds or brass), our system will create the full section. For example, a recording of the first violin and cello will provide you with a full string section, including one divisi. For most media music scores you will probably need to record only six musicians, but for complex scores you may need up to 12 musicians.
This raises the question: which parts to record, and which to leave as virtual instruments?
The answer lies in the nature of the musical expression of a given part, and the way it relates to the other parts. In general, orchestral musicians either play with musical expression that is similar to other instruments, or with musical expression that stands out from the rest of the orchestra (a ‘solo’).
In our system, any instrument that is played with the musical expression appropriate for other instruments can, for that period of time, be the Expression Leader for the other instruments.
Conversely, when a musician plays an independent line (a ‘solo’) it would be inappropriate for other instruments to follow his or her soloistic playing. In this case, the part must be recorded by a musician.
To summarise, there are two golden rules to follow in order to get the best stems for your score:
It is frequently and easily said that every musician is unique. But it is important to fully appreciate the significance of this fact. We have tested and demonstrated that every skilled musician paired to an instrument has a musical signature that is unique to that particular pairing.
Ask a different musician to play the same instrument, or ask the musician to play a different instrument, and in both cases the musical signature will change. In order to calibrate our system with the unique musical expression of each recorded musician, we need to create a profile for the musician. This entails recording a few scales in a wide dynamic range (parts can be downloaded here).
Once created, a musician’s profile can be stored in our database, and as long as the musician plays on the same instrument, the musician can be hired for subsequent sessions without requiring a new profile.
You can find more details on how this works in the Musicians’ Profile tutorial.
We require close-mic recording methods that enable us to capture the full range of expression of musicians, especially when they are playing in a group. We use industry standard recording equipment, and full guidance for how to record all instruments can be found here. Even if you are an experienced engineer, please take the time to review this guidance to make sure you have the full information needed to get the best results from our service.
The complete process is quite simple and consists of four steps.
Step 1: Score preparation
You prepare your MIDI score, focusing on composition and orchestration. As mentioned earlier, you will need to identify the parts played by your Expression-Leader musicians. As a reminder, here is the link for the relevant tutorial, and information regarding project setup, can be found here.
Step 2: Record
Your musicians must record their musical profile, if they have not already done so. And of course, your musicians will record the selected parts of your score. As always, it is important to ensure your musicians record in a manner that produces material which is coherent in timing and expression, even if there are only a handful of players, and especially if they are doing so remotely from each other.
When you prepare for the session, please take the time to become familiar with our recording method (see here for the How to Record tutorial).
Step 3: Upload
In this step, you upload three groups of files:
We will send you an email to confirm that we received the files, or to let you know if there are any problems. Full details on how to send us your project can be found here.
Step 4: Download
Once we accept your files, processing of your score can take up to 24 hours. We will send you an email when the stems are ready for you to download.
We offer three different packages, each providing you with a variety of standard mic positions and instrument groups. Depending on the package you choose, you will get between 6 and 65 stems, with levels balanced to give you a starting point for your mix.
For details, please review our full tutorial, which also gives examples of the different stem-package outputs. On our download page, you will find full sets of stems (and the scores) for each of the packages for you to test mixing. You can also listen to examples here.
LOGISTICS & COST
On our website you can also find a list of collaborators who are familiar with our system and who would be delighted to work with you. You will find:
Please do get in touch here if you’d like to become a collaborator.
This is version 1 of our new service. The core technology was developed over many years to be used in live performance, and we are very excited to listen to what creative composers will do with it.
There are a number of capabilities and features that will be added in upcoming iterations and releases. Feel free to suggest anything you think we should include that would make your work easier or better.
We hope that you’ll enjoy using our system, and look forward to listening to your productions!