You are here

Steinberg Wavelab

Digital Recording & Editing Software By Janet Harniman-Cook
Published August 1996

This sophisticated new PC digital editor is up against such established software as Sound Forge and Samplitude Studio. JANET HARNIMAN COOK assesses whether Steinberg have built on their status in the sequencer field with a successful PC wave editing program.

As the PC music juggernaut gathers pace, the proliferation of 'MIDI + Audio' sequencers and budget hard disk audio recorders continues unabated. Software‑driven, and using soundcards within the PC, these programs offer a fantastic range of creative opportunites to the musician of today. But although they may be great multitrackers, the audio editing facilities of these programs may be rather basic, often restricted to simple cut, paste, volume and pan functions. To provide the missing editing power, Steinberg have launched WaveLab. To help you reach that elusive next level of refinement and control, WaveLab provides a bank of fast audiophile wave processors, including Timestretch, Chorus, Harmoniser and EQ with spectral analysis, all integrated into an easy‑to‑use, high‑quality environment for all those tricky edits. If you're on a tight budget, you could just about use WaveLab as a stand‑alone recorder for compiling albums or working on extended remixes, despite the limited playlist features — sadly, WaveLab is not yet the PC's answer to Sound Designer II on the Mac. For the multimedia developer, WaveLab provides fast, high‑grade sample rate conversion and batch processing.

WaveLab is two program modules: a stereo digital audio recorder/editor/multi‑processor, and the Audio Access database, a disk‑orientated soundfile librarian. It works with any 16‑bit, stereo, 44.1kHz Windows MME soundcard, both PC WAV and Mac AIFF file formats are supported, and WaveLab will edit Digidesign Session 8 files too. The program is also ideal for use with the current version of Cubase. Audio files from Cubase v3 (see review in the July issue of SOS) can be directly imported into WaveLab for fine editing and processing, and Cubase v3 audio segment boundaries are translated into WaveLab markers.

Package Deal

The package consists of a CD‑ROM and a clearly‑written manual with comprehensive index and tutorials. The CD‑ROM contains the program, Cubase information, and over 200Mb of wave files from Best Service, including some utterly gobsmacking drums and instruments from New World Order Journeys 1 & 2 (watch out for a full SOS Sample Shop review shortly). Copy protection relies on occasional polite requests to insert the CD‑ROM — so much more civilised than the dreaded dongle! WaveLab has Windows Help and on‑line contextual Help, and there's also 'Tips' Help, which pops up when you point to a screen icon, and describes its function. Tip of the Day consists of a Help box containing a brief explanation of a WaveLab routine or shortcut. All in all, it's a very useful set of reference and learning tools.

From the moment it boots, WaveLab looks special. Configurability is the name of the game. The colour of most text and graphics is user‑definable and can be saved as a Style in the Options menu. Most windows have 'speed' menus, activated by clicking the right mouse button, which contain the most commonly‑used functions — you can quickly switch between the units used for Time and Level measurement using the Rulers speed menu, for example. WaveLab also features a comprehensive set of key commands to help speed up operation. Multiple files can be opened and displayed, and you can save your workspace on quitting; when you next boot it up, WaveLab will automatically load WAV files and database information, and even recall a Toolbar position. Projects are the best way of working in WaveLab (unless you're doing only simple, one‑off edits) and these also retain Snapshot and Marker information. All editing in WaveLab is non‑destructive to your source recording, unless you choose otherwise.

Getting To Work

The WaveLab Applications page contains the Menu and Status bars, wave views, Toolbars and, when open, the Audio Access database. The Status bar at the foot of the applications page displays wave cursor position, zoom factor, wave/selection length, and sample parameters. During processing tasks, progress times are also shown. Toolbars can be floating, or 'dragged and docked' along the sides of the application window. Toolbars may be hidden and recalled as desired, and transport functions have their own Toolbar, as does the Toolbox, which accesses selection mode, scrub, and the inevitable Steinberg kickers (nudge tools). The Control bar icons access the more common File and Edit functions.

The wave windows consist of two panes: the Overview and the Waveview. The Overview, as its name suggests, shows the whole soundfile, and is mainly used for information and locator functions. Clicking anywhere on the Overview locates the wave window cursor. To define the area of the file you wish to work with, you create a selection by dragging the mouse over the wave display. An area changes colour to indicate the selection. Selecting an area in the Overview defines the portion of the soundfile displayed in the Waveview window beneath.

The more I use WaveLab, the more I like it.""Editing in WaveLab is a joy — procedures are fast and well thought out.

The bulk of WaveLab activity takes place in the Waveview pane, where you can work on both channels independently, or on both, as a stereo file. Wave displays can be magnified or shrunk, using the horizontal and vertical zoom sliders found by the scroll bars. The zoom sliders can be dragged with the mouse, enabling you to view the whole file and to zoom down to one sample per pixel, if required. To avoid time lost scrolling around the file and changing magnification level, WaveLab allows the same audio data to be viewed in multiple windows. Each can have different ranges and zoom settings. In addition, each wave window can store eight 'Snapshots'. These are Waveviews containing range and zoom information, which can be saved, recalled, and overwritten. Instant switching between windows during playback is possible by clicking on the time ruler of an inactive window. This technique is useful for comparing audio — for example, when sorting between takes.


Markers are used as locators or section labels, and appear as inverted triangles on the Waveview measure ruler. They are created — during playback if you wish — by pressing Insert, and they can be repositioned by dragging, and even given names. You can view the Markers list by double‑clicking on a time ruler or using [Control] M on the keyboard, and the list can be sorted by name or position, but curiously there is no provision to edit Marker positions numerically or to snap Markers to time. Positioning the mouse beneath the Marker icon makes the Marker name appear in a 'Tips' box. This system works reasonably well, but Marker names can only be seen one at a time. I would prefer a hideable Marker bar, containing all the Marker names, beneath the time ruler.

The wave cursor is located to the Marker position by double‑clicking on the Marker triangle — this will even work during playback. If the Marker is not visible — for example, if it is outside the wave window's current range — clicking on its entry in the Markers list produces the same result. Double‑clicking on the waveform between two markers selects the audio between them.

Other navigation aids include the 'Go To' options accessed from the View menu or the Waveview speed menu. Clicking on the cursor field in the Status bar locates the Waveview to the current cursor position, and you can cue the audio by clicking on the Waveview time ruler. The Transport bar contains the loop selection, record and stop icons: clicking once on the Transport Stop icon locates the cursor to the start of the selection; clicking twice sends it to the beginning of the file. The numeric pad also contains locator functions, and you can position the cursor to an exact time by using 'Move Cursor to Position' in the View menu.

Editing Operations

Editing in WaveLab is a joy — procedures are fast and well thought out. Simple edits can be made by drag‑selecting an area in the Waveview. The Waveview speed menu offers options including Cut, Copy, Paste, Delete, Replace with Silence, Play Selection, Zoom Selection, Screen Elements and Colour Preferences. As WaveLab is truly multitasking, you can edit and process while looping playback. This allows you to quickly audition your edit and undo changes on the fly. Impressive stuff — although you do need a powerful computer for glitch‑free playback during the more demanding edit processes, such as complex chorusing and harmonisation.

Sections of audio can be copied to a different position within the soundfile using drag‑and‑drop, and if you hold down [Alt] and drag, the audio will be moved rather than copied. To create a new wave window, all you need to do is drag your selection into the main application area. You can copy between soundfiles by dragging your selection into another wave window, although take care not to drop onto any selected audio, or you'll create a crossfade. Other basic edit functions include Trim, which deletes all audio outside the selection, and Insert Silence, which will place a blank section the length of the selection into the soundfile at the selection start point. You can also convert a mono file to stereo (and vice versa), and flip the channels of a stereo file.

With the 'Magnetise Bounds' command from the Options menu active, you can use a Marker as a snap point. Just drag the audio near to the Marker line, release the mouse, and the audio will be positioned at the Marker point. If Snap to Zero Crossing is active, WaveLab will automatically find the best place to create a glitch‑free splice. Audio can be shunted to the beginning or end of the soundfile using the Prepend and Append commands. Another useful option is 'Multiple Copies'. With this, you can extend that killer drum loop to create a temporary working drum track — up to 1000 copies can be created in this way. Two sections of audio can be blended by copying one to the clipboard and then using a 'Mix' command to combine the two.

Wave Processing

WaveLab's wave processing functions divide into two main areas: level correction and effects processing. During processing, the Status bar shows task progress times, which were generally surprisingly fast on my Pentium 100 system. Once processing is complete, you can use Undo to return to a previous version. WaveLab has multiple Undo levels, and each level is consecutively displayed on screen. While this is invaluable, I would prefer an Undo history, listing edit operations with time and file details.

WaveLab supports batch processing. This is a very powerful feature that enables you to process several files using multiple treatments in a single operation. You might use batch processing to correct DC offset (more shortly on this), or to normalise, compress and gate different takes from a recording session, prior to editing and assembling the final version. Beware, though — batch processing can be relatively time‑consuming, even on a fast PC. Large files — especially using multi‑stage processing — can take a long time. The best thing to do is organise your task list and go and make a cup of coffee.

The most often used Level functions are Change Gain and Normalisation. Change Gain adjusts absolute volume — that is, relative to the level of the audio itself — while normalisation adjusts volume level relative to 0dB, the threshold of digital distortion. Normalise is most commonly used to optimise the level of soundfiles recorded with low gain. Both Change Gain and Normalise dialogues also include a 'Get Peak' level option. The DC (Direct Current) Offset function, under the Level menu, allows correction of DC Offset problems, which are most commonly experienced with soundcards and some DAT recorders, and can be heard as clicks at splice points, caused by mismatches between digital audio recording devices. The problem is part of the recording process and can be seen if the wave image appears off‑centre from the zero level axis when you zoom into a quiet part of the soundfile.

The Level menu also includes Fades, and the Fade dialogue box contains two parameters: Damping adjusts the steepness of the fade, and Offset defines the point at which the fade reaches ‑6dB, or half the volume level of the soundfile. The Fade curve appears superimposed on the Waveview, with a small square representing the Offset focus. The review copy of WaveLab contained a bug that occasionally displayed the fade curves incorrectly, although processing itself was not affected.

WaveLab includes sophisticated dynamics processing. Compression, expansion, limiting and gating are edited from the Dynamics dialogue by adjusting the familiar envelope parameters of Attack, Hold, Release, Threshold and Ratio. (For a detailed look at compression, see Paul White's article in SOS April '96.) You can also use a graphic editing option, and presets have been included by those clever Steinberg boffins to emulate hard‑knee, soft‑knee and gated tube compression. There are some very fine algorithms at work here that added punch and body to test pieces I ran through the presets. The bass and lower mid frequencies were given enhanced depth and crispness, while the top sat more comfortably in the mix. Had I spent £400 on a dedicated rack unit that produced results of this quality, I would have been delighted.

The timestretch algorithms in WaveLab are amongst the most natural I have ever heard. Timestretch is most often used to slightly speed up a finished track to give it more 'wizz', or to synchronise two pieces of audio with different lengths or tempos. From the Timestretch dialogue, you can specify the degree of timestretch required as bpm (beats per minute), time, samples, ratio, or SMPTE timecode. It is usually on complex, wide‑bandwidth recordings that the shortcomings of timestretch routines are most obvious. I chose a full‑range mix of vocal, drums, bass and organ, with an original tempo of 108bpm. I shifted the material, whilst retaining its true pitch, first up to 120bpm and then down to 96bpm. Processing was surprisingly fast, and the quality of the results was impressive, with very little evidence of aliasing.

Pitch Correction allows you to change pitch, with the option of altering length. The WaveLab Harmoniser can create up to 16 shifted voices, which can be individually mixed for pan, level and pitch. The Harmoniser is also very useful for creating stereo from a mono source. If you use it subtly, you can avoid the 'munchkinisation' that occurs when simple pitch‑shift devices are applied to the human voice and acoustic instruments. Generally, pitch‑shift‑type processing is best applied in small increments, except when used for special effects and synthesized sounds.

The Hi‑Fi Chorus is tasty. WaveLab copies the source signal up to 100 times, and just as in a real chorus, each voice is slightly detuned and delayed. The results are very sweet‑sounding. Parameters include number of voices, pitch variation, delay, dispersion, and intensity. Beware, though, if you do choose a multiple voice chorus: you may have to wait a long time for processing. Steinberg development, please take note: for short operations, WaveLab processing and Undo is fast enough, but for long, complex operations WaveLab needs a Preview function.

Frequency Analysis & EQ

WaveLab has a rather snazzy Frequency Analysis function which allows you to view audio by its timbral content. Having made a selection, choose 'Do Frequency Analysis' from the View menu. The audio is displayed as a 3D Fast Fourier graph, with amplitude represented by rainbow‑coloured peaks and valleys. Right‑clicking on the graph calls up a dialogue box from which you can change your viewpoint, choose a Linear or Logarithmic representation, or different colour options. I've always been fascinated by 3D FFT graphs and this one is gorgeous! From a more utilitarian angle, Frequency Analysis is a vital tool for identifying rogue frequency components in audio signals. The only small gripe I have is the lack of user presets. When using Frequency Analysis on a test piece, I identified a high‑amplitude, sub‑bass component at around 23Hz, which I'd failed to hear on my monitors when recording, because the frequency fell outside their accurate working range. Sometimes low frequencies like this can cause problems, so I wanted to hear the audio with the sub‑30Hz band removed. Unfortunately there was no way I could do this using the WaveLab EQ (which has two shelving filters — high and low — and a parametric, wide‑range mid band with adjustable Q), as 23Hz falls below the frequency of the parametric (50Hz‑14.6kHz), and the fixed curve of the low‑shelf filter (45Hz‑2kHz) attenuated higher frequencies that I did not wish to alter. Similar difficulties arise at the other end of the audio spectrum, as the top filter has a fixed Q and a range of 2kHz‑12.3kHz. For most purposes, the WaveLab EQ will be adequate, but the addition of low‑ and high‑pass filters or a wide‑band graphic equaliser in a future upgrade would be most welcome.


So how does WaveLab compare with its rivals in the race for your money? Its processing power would certainly be enhanced with the more comprehensive EQ, multitrack facilities, and real‑time preview functions of Samplitude Studio. And it would be great if WaveLab had the wide processing palette of Sound Forge v3, or the MIDI and SCSI sample dump, reverb, and EDL playlist tools that are part of both of these admittedly more expensive programs. I'd love to be able to escape the tiny edit LCD on my sampler, and I think it's a pity that WaveLab does not support MIDI or SCSI sample dump. However, WaveLab has unique features that Samplitude Studio and Sound Forge do not have (or which are only available as plug‑ins): Frequency Analysis, batch processing, amazing user configurability, and the Audio Access database. In addition, there is an undeniable elegance about WaveLab editing and processing that no other PC software quite possesses. WaveLab just oozes quality, and is very, very fast. It brings the best out of Windows 95 and presents a clear and refreshing graphic user environment which, when combined with the flexibility and ease‑of‑use of the WaveLab editor, provides a high‑quality workspace for the engineer. The more I use WaveLab, the more I like it. Highly recommended.

Many thanks to Michael, Simon, and Steve Peterson and the gang at Look Micros.


A Playlist, or EDL (Edit Decision List), determines the play order of tracks or part‑tracks and is used in soundtrack work, CD compilation, and remix sessions. The EDL typically contains start and end times and duration information. At first glance, WaveLab does not have an EDL, but fear not — there is a way around this. Load your tracks into WaveLab as usual. Create a new blank wave window: this will be the master track. Copy your tracks in sequential order to the master track, adding markers denoting start and end points as you do so. Double‑click on the time ruler to open the Markers view and sort by time. The start and end points of the tracks will be presented in order. Use the WaveLab editing tools to edit the list.

System Requirements

WaveLab v1.01 is a full 32‑bit native Windows 95/Windows NT application and will not run under Windows 3.11. The recommended requirements are a PC with maths co‑processor (386/486DX or Pentium) with a minimum of 8Mb RAM. In practice, I would recommend at least a 486 DX4 100, and the more RAM the better. As with all hard disk recorders, you need a fast, large SCSI or E‑IDE (mode 4) hard drive. A CD‑ROM drive is also required. Best performance will be achieved using the program in 256‑colour mode.

You'll need to change the Windows 95 Virtual Memory settings in System Properties/Performance, from the Control Panel. Don't be deterred by the dire warnings — Windows 95 can sometimes be unnecessarily dramatic! Specify 20Mb minimum and between 20 and 40Mb as the maximum. The optimum size is two and a half times your amount of RAM. You can also save your WaveLab virtual memory configuration as a separate hardware profile in System Properties, should your WaveLab configuration conflict with other software.

Review System: Pentium 100 PC; 256K pipeline cache; 16Mb RAM; standard PCI video card; 17‑inch monitor; Turtle Beach Multisound Classic soundcard for analogue; and DAL Digital‑Only Card D.

Access All Areas?

WaveLab contains a proprietary database called Audio Access. With it you can catalogue, locate, audition or transfer to the WaveLab editor any soundfile you have stored on hard, removable or floppy disks and CD‑ROM. You can organise your soundfile library into categories and groups, so that similar types of soundfiles from different sources are displayed together. For example, your Brass group may be sub‑divided to include a trumpets section, which in turn may include riffs, loops, ensembles, single‑note sets, and so on. You can use the database search facilities to sort and display files that meet criteria such as instrument, date, size, sample rate, length, or user keyword. Similarly, you can set filters to exclude files that meet defined criteria. As a means of getting to grips with WAV and AIFF files, the WaveLab database is great. But a large proportion of my digital soundfile library is on audio CD or DAT, and sadly there is no provision for these. However, Audio Access is a studio tool that will become increasingly more useful as I transfer my old sample library over to CD‑ROM and hard disk.


  • Exemplary editing environment.
  • Excellent audio processing quality.
  • Very fast.
  • Audio database.
  • Frequency analysis.
  • Stable and reliable.


  • Needs at least a DX2‑66 PC.
  • No preview function or real‑time processing.
  • No EDL playlist or sampler support.
  • No reverb or DDL.


WaveLab is a powerful, classy program with a modest price tag, and will boost the wave editing power of most 'MIDI + Audio' sequencers and budget hard disk recorders.