The human voice can have a dramatic emotional impact. Unfortunately, too many voice-based sample CDs are mired in clichs and fail to reach for broader

The human voice can have a dramatic emotional impact. Unfortunately, too many voice-based sample CDs are mired in clichés and fail to reach for broader creative possibilities. Voice Fusions ($99.95, audio CD), from the fledgling Soundforest company, makes that risky reach with distinctive results.

Speaking of Which

The self-described “explorations with voice” travel a diverse aural landscape, often pulling off on the sonic back roads to see what's there. Lurking in the two-CD set's 190 tracks is everything from spoken phrases to complex sound-design transformations; hundreds of sounds provide a lot of bang for your buck.

Producer Michael Kelsey created the sounds from three basic stylistic ideas: voice synthesis (using the human voice as waveforms for an E-mu E-Synth and Kurzweil K2600); language as music (metaphorical or unusual, thought-provoking phrases); and voice explorations (the suggestive nature of the sample itself inspiring new phrases or rhythms).

Kelsey then organized the sounds into 31 categories according to “a predominate impressionistic quality,” with titles such as Dream Deliriums, Poetry, Voice Rhythms, Shaman Signs, Rubber Springs, Granulation, and Alien Worlds. Samples run from one second to more than two minutes and are arranged from shortest to longest on each track.

Fuse Box

The material is oriented to the spoken word, vocal effects, and vocal improvisation as opposed to traditional singing. Vocalists were chosen for their ability to project their personalities rather than those of the microphones (Røde NT1s, a Shure SM57, and a Radio Shack PZM) and for their ability to enunciate, accent, and inflect. Kelsey kept an ear open for happy accidents. I have a hunch that there were happy accidents aplenty.

The typical “ooh,” “ahh,” and “oh, baby, do it again” samples are absent, though Voice Fusions does include shorter statements that could work in a dance-music context. Instead, you'll find surreal, abstract phrases such as “a clock surrenders to the filaments of light surrounding it,” and “I am not unhappy, only wet with darkness, soaked to the bone with primitive memories.” Many phrases were excerpted from poetry written by Kelsey; some are stream of consciousness musings.

Kelsey took a no-holds-barred approach to the processing, using Steinberg's WaveLab, Sonic Foundry's Acid, Waves' Gold Native Bundle plug-ins, and hardware units such as the Ensoniq DP/4. You can hear standard effects such as delay, panning, vocoding, pitch shifting, and reverb in the collection. However, at least half of the material has an intuitive, improvised quality thanks to digital signal processing that follows the flow of the sound, completely altering or enhancing it. Many samples use multi-effects processing. Although some sounds are composites of two or more samples, most additional textures and artifacts are a result of the processing.

Intriguing Utterances

I discovered numerous gems in addition to the poetic and stream of consciousness musings. The eight samples in the Old Time category create an antique radio atmosphere; in fact, the sample Tonyboy was created from family recordings dating from the 1940s and 1950s. Monkschant and Mountainchant are lengthy, evolving sounds, each evoking the aura of their names in oblique, dreamlike ways.

The Voice Rhythms and Beatland sections contain rhythmic material that you can loop. Clips such as the entrancing Velvetmorn and Enigbeat are just the thing for trance music; the Chickenmeeting samples are edgy with spiked rhythms that are suitable for a more industrial setting. Zeiglig2, from the Syllable Play category, has no bpm reference but is an enigmatic and cool rhythmic bit.

The samples seem to flow from shorter, more intelligible sounds at the beginning to more intensely processed and experimental things. The Humangeese sounds on disc 2 are superbly designed abstractions, as are Forcesalittle, Indecision, and Alcoholcontent from the Scitter-Scat section.

The last few categories offer the longest and most bizarre designs: deep, powerful, primitive growling; shimmering, doppler-esque atonal singing; twittering ambiences; futuristic space rides; and on and on. Walkaround2 in the Transformations category is a wash of cacophonous noises featuring a discernable spoken voice that emerges occasionally. Itsaknow evolves over a period of 38 seconds in a style reminiscent of Steve Reich's “It's Gonna Rain.” It's hard to believe that Sounddeepcrystal, an evocative two-minute drone, is of vocal origin.

Voicing My Opinion

A unique, sometimes strange, and often inspiring collection, Voice Fusions is an admirable initial offering full of wonderful surprises. The human voice can indeed be dramatic, but so can an adventurous approach and attitude to recording and processing.

Kelsey suggests using the collection in contexts that include mainstream music, experimental music, electronica, choreography, film, theater, and performance art. The sky is the limit, though. One rewarding approach is to arbitrarily cut and paste a few phrases together as though they were refrigerator poetry magnets.

The documentation covers the essentials, though the indexing system has a few errors in the length of the samples. You can download a more comprehensive, sortable database that includes bpm and byte-size listings from Soundforest's Web site. Those who appreciate truly artistic sound design must give Voice Fusions a listen.

Overall EM Rating (1 through 5): 4.5

Soundforest; tel. (877) 237-5771; e-mail; Web