Manifest Technology
        Making Sense of Digital Media Technology
        By Douglas Dixon
 - PC Video
 - Web Media
 - DVD & CD
 - Portable Media
 - Digital
 - Wireless
 - Home Media
 - Technology
     & Society
 - Video - DVD
 - Portable
 - What's New
<<< HOME 



  Manifest Technology Blog -- Site: | Articles | Galleries | Resources | DVI Tech | About | Site Map |
    Articles: | PC Video | Web Media | DVD & CD | Portable Media | Digital Imaging | Wireless Media | Home Media | Tech & Society |
    PC Video: | PC Video Articles | Video Software Gallery | Video Editing Resources |

Adobe Soundbooth CS4 Beta: Scores  (7/2008)

    by Douglas Dixon

Introducing Soundbooth CS4
Soundbooth Scores for Music and Sound
Speech Transcription
XMP Metadata

See Adobe Soundbooth CS3: Audio for Video for a walkthough of Soundbooth's interface and features

See Adobe Audition 3 -  Professional Audio Editing and Mixing for more on Audition for professional editing
and Summary: Adobe Audition 3 for Audition comparison to Soundbooth, new features, features summary

See Adobe Creative Suite 3: Applications, Suites, New Features for an overview of the CS3 applications, suites, and pricing

Adobe Soundbooth is the audio editing component of the Adobe's Creative Suite collection of design and development tools for print, video, and web (



Of course, Adobe's suite of tools, now Adobe Creative Suite 3 (CS3,, is best known for immensely powerful applications for professional users -- design tools including Photoshop and Illustrator for image and graphics design, production tools including Premiere Pro and After Effects for video and motion effects, and web tools including Flash Professional and Dreamweaver for interactive animation and website design.

These are the kinds of tools that users immerse themselves in almost full time as the core focus of their work. Soundbooth, however, is designed for a different purpose. It is an ancillary tool, designed to help creative professionals get in and out quickly to accomplish audio tasks -- cleaning and enhancing clips, and generating quick audio mixes -- to sweeten a video production or Flash animation.

Adobe first added a separate audio tool to its product line with Adobe Audition (, introduced as part of the Video Collection suite in 2003, based on the Cool Edit Pro product acquired from Syntrillium Software. But as Audition was enhanced as a power tool for audio professionals, it became less accessible for video creatives. As a result, Adobe introduced Soundbooth with Creative Suite 3 in 2007, and split Audition out of the suite as an independent product with its own product development cycle.

Soundbooth has a list price of $199, compared to $349 for Audition. 

     Find Adobe Soundbooth CS 3 (Windows or Mac) on

     Find Adobe Creative Suite 3 Production Premium  (Windows or Mac) on
     Find Adobe Creative Suite 3 Master Collection  (Windows or Mac) on

Introducing Soundbooth CS4

Now, as Adobe gears up for the next CS4 product cycle, it has posted prelease beta versions of several of its applications on its Adobe Labs site, including Soundbooth, Dreamweaver, and Fireworks ( 

The key features of the new Soundbooth release include:

  • Multitrack support
        Mix across multiple clips to create layered soundtracks, including Scores
  • Volume matching
        Equalize individual clips, Match volume levels across multiple clips
  • Beat detection
        Automatic, tempo-based beat detection for creating loops
  • Preview MP3 export
        Preview MP3 compression before saving, balance bitrate and file size
  • Enhanced Soundbooth Scores
        Choose from Score library, on-track editing in multitrack view
  • New Adobe Sound Document (.ASND) file format, with Snapshots
        Save multitask project with source clips, edits, mix
        Save multiple instances of project as Snapshots -- with all edits of clips and tracks
  • XMP metadata collection, including text-to-speech transcription

So you can download the Soundbooth CS4 beta, which will run for 30 days, or until the product is released if you have a CS3 serial number (

Let's highlight some of these new features that you can explore in the beta download.

    Soundbooth CS4 beta: Multitrack mix - 2 music clip waveforms and 2 scores with parameters
        Left column: Scores panel, Properties panel, History panel

Soundbooth Scores for Music and Sound

Soundbooth Scores are used to create professional-sounding royalty-free music beds, similar to SmartSound Sonicfire ( Soundbooth's scores are available in libraries, with genres like Wedding, Atmospheric, Country, Jazz, and World. These products typically ship with a core selection of scores, with more available for purchase.

The original idea with these tools was to help you create music to match your production without having to manually cut and paste finished pieces from a stock music library. Once you select a score, the software can automatically match the exact duration on your timeline -- the scores are designed in segments that can be assembled and adjusted like blocks to fit the desired duration, flow cleanly through different movements, and still have a natural intro and ending.

Even better, the scores can be tweaked to not always sound the same -- in Soundbooth you can select a variation with a pre-defined mix of segments, and choose to include or omit the Intro and Outro segments.

Beyond adjusting the segments horizontally by duration, the next step with these automatic scoring tools was to add more control by adjusting the mix vertically -- breaking down the score as a multi-track mix. Soundbooth scores provide an Intensity parameter that controls the depth or complexity of the music, adding more layers of sound. Individual scores also provide additional parameters that fit the musical style, such as emphasizing the Lead / Background, Solo for a Jazz piece, or highlighting the Melody or Piano.

    Soundbooth CS4 beta: Score track maximized, with full parameters and keyframes
        Score Properties panel on left

With these controls, you're effectively mixing a multi-track composition, but in a way that's part of a musically coherent whole. Even better, to further customize the feel of the music to match your production, you can add keyframes in Soundbooth to adjust the mix, and the volume, on the fly during the production -- for example, adding depth to match the action or simplifying the mix during dialog.

With Soundbooth CS4, Adobe has taken the score idea a step further by adding a library with a new kind of scores -- Environment scores that are multi-track sound effects. Instead of automatically generating music in a particular style, you can use scores to generate sound effects to add environmental texture to your mix, and again assembled from blocks to fit the desired duration.

And the environment scores are also constructed as multi-track mixes, so you can adjust the Intensity parameter to layer in more elements as desired. For example, with the CityStreet score, Intensity controls the amount of traffic driving by, Construction adds noise from trucks, plus Rain mixes in drops and thunder.

Soundbooth Scores provide a quick and efficient way to add background music to a video production to help develop the general mood or feel of a piece. And the new environmental scores in Soundbooth CS4 extend this idea further for ambient sound design to match your visuals. Then use the new multitrack mixer to add additional tracks of music and sound -- layers of layers of texture for your production.

So download the beta to try it out (which runs fine with Soundbooth CS3 still installed), and then we can look forward to the libraries of music and sound scores that Adobe provides with the product when it is released.

Speech Transcription

But there's another less-obvious addition to Soundbooth CS4 beyond editing individual clips -- a focus on collecting and tracking metadata associated with your clips, to make it easier to search and manage your growing collection of clips and projects. As a bonus, Soundbooth now supports speech-to-text transcription, to generate a transcript associated with a clip -- making it easier to search to find a particular clip, and to shuttle though clips to find a specific edit point.

The transcription process is not magic -- Speaking generally, it works best with professional-recorded broadcast-quality clips, and less well with more challenging clips, but there's no really precise specifications or expectations at this time. The product will support multiple language models, including variants such as American or British accents, to help improve the accuracy (the beta just includes U.S. English). It also provides options to try to identify multiple speakers, and to select faster or higher-quality analysis.

But, argues Lawson Hancock, product manager for Soundbooth in his Inside Sound blog, it's not just about the accuracy of the transcription (

... "Even with a 50% accurate transcript, the speech engine will capture enough keywords to enable you to easily search through your audio stream. ... Speech search is about speeding up the editing process by enabling you to easily locate specific words in your audio clips."

As a result, the engine is tuned to err on the side of always finding words, even if it has to guess at a lower confidence level (which Soundbooth displays for each word). You also can edit the transcript in the Metadata panel to correct specific words as needed.

    Soundbooth CS4 beta: Clip (waveform) editor
        Metadata panel on left, with Transcript at bottom with current word highlighted

As you play and skip though the clip, the text in transcript is highlighted accordingly, so you can edit the audio at a more helpful level of words and sentences, instead of working with samples and waveforms.

The real bonus comes over time as you build a collection of clips with transcripts -- the words are associated with the clips as metadata, and are searchable and usable in other tools when you need to find a specific clip or even exact quote.

XMP Metadata

The transcript is only part of the extensive metadata collected and managed by Soundbooth CS4, and shared with other Adobe applications. The idea is to capture information about clips as soon as they are imported or created, and then augment and share the data though the production process.

Based on Adobe's Extensible Metadata Platform (XMP,, data about each clip is embedded and stored in the file itself when saved, and then displayed in the Metadata panel. (Some file formats including AIFF & MPEG do not support embedded XMP metadata, which is instead saved in a separate .xmp file stored with the file.

The metadata stored by Soundbooth includes file attributes, rights management, and extensive audio properties even including Flash cue-point markers and Beat markers.

Again, the power of this metadata then grows over time, as you can quickly find specific clips by searching for very detailed attributes. We can move beyond organizing files by folders, or date and project, and start to think about them in terms of how they were created, processed, and saved. This will take new thinking and new tools, but we can already see the promise in Soundbooth CS4, with XMP metadata and text transcripts.


Adobe Labs

Adobe Soundbooth

Adobe Audition

Adobe Creative Suite

SmartSound Sonicfire

Adobe Extensible Metadata Platform (XMP)

Adobe Soundbooth is apart of the video Adobe Creative Suite 3 Production Premium -- including Premiere Pro for video editing, After Effects for motion graphics and visual effects, DVD production with Encore, and now Soundbooth for audio editing.