StableAudio.com Test #7: Americana: KHÉYA WÍTA (Got good headphones? Use 'em!)

KHÉYA WÍTA (Turtle Island)

As part of the “AI Americana” series of tests, I planned to do one for Native American music. I did not expect what happened, and the results below. Again, I take no credit, nor is this TATANKA. It’s AI and credited as such. But it is unique and worth a listen. TATANKA will not use AI for writing, inspiration or performing of any music, so unless The Council tells me otherwise, this will be the only version. Links to all are below, including an MP3 of the single-track 71:04 minute “mix” if you want to easily grab a free copy.

This is not cultural appropriation. This is respect, and love of Turtle Island‘s (original name for North America – “KHÉYA WÍTA” in Lakota – Google it if you didn’t know) indigenous cultures and wisdom, to which TATANKA is aligned (ᗑᑌᔑᓵᐸ ᗑᗕᗕᐪᔑ ᗑᓵᔑᔑᓵᗝᐱ™), and aspires to bring awareness, as it is critical to the sustainability of humanity.

I used the following text prompt:

“Great Plains (Sioux) Native American Pow Wow song, leather head drum circle, hand drums, frame drums, native american flute, bells, rattles, people whistling, the sound of feet in time as people dance to the music, male and females chanting, singing. The voices are the melody and drive the music. Key of F#. BPM 80. 6/8 time signature.”

Usually I run a prompt a few times and select the best result, but each version generated was compelling enough that I just could not decide as each is fundamentally different (I was surprised at the creativity and diversity AI illustrated) and just as beautiful. So I decided to run twenty-four versions (I kind of get into a groove), compile them, and post them together, like a traditional album. None of the tracks are traditional. All are derivative, and some venture into trance-like, trippy tuneage. In other words, listen with open mind, heart and ears.

StableAudio.com was wonky today, not allowing me to download finished files, so I had to use a screen capture app. Since the browser “video” was also captured as a result, it was not a flawless stream, and occasionally a minuscule pop is heard. Odds are only my fellow audiophiles will notice, so let’s call it a “Vinyl Rip, with Skips and Scratches.”

Right.

I mixed all the tracks to overlap at the “heads and tails,” in sync, so the result is one compiled 71:04-minute “song” with twenty-four parts. Since each part, song, is the same key, tempo and meter, it was easy to fuse all into one continuous track.

One of the cool aspects, to me, is although everything is digital, the final versions are faux analog due to my computer issues resulting with the “vinyl effects.” I think that’s fitting.

I thought it might end best on a strong note, a climactic cacophony, so I built up to it a bit. I know it’s cliche, but I visualized a bow, pulling back an arrow, then letting it fly to find its target, so #23 is scaled down, a wind-up, and #24 is the fastball. I threw it in the sweet spot, so take out those center field lights!

Track #23 prompt:

“Great Plains (Sioux) Native American Pow Wow song, leather head drum circles, hand drums, bells, rattles. Simple, rhythmic, but the drums powerfully drive the song. Key of F#. BPM 80. 6/8 time signature.”

Track #24 prompt:

“Great Plains (Sioux) Native American Pow Wow song, multiple THUNDEROUS leather head drum circles, pounding hand drums, powerful frame drums, numerous pounding, different wood percussion instruments, native american flutes, various bells, various rattles, male and females SHOUTING, yelling, chanting. The loud human voices drive the music. An indigenous wall of sound and fury. Key of F#. BPM 80. 6/8 time signature. Fade out and end with a warrior’s cry.”

The separate tracks are individually available on the TATANKA bandcamp page, in addition to the “album track” but make sure you listen to the full mix in one sitting and yes, if you indulge, this is a fitting soundtrack. Worth mentioning is the evolution of AI creativity, track after track, specifically the chanting and singing. For example, in track #11, there is a sudden and powerful moment that comes and goes, not out of context, but certainly in contrast to the track, some might argue… improvisational? (Don’t get me started but yes, I believe we are already in AI 2.0, the post-singularity, applied AI adolescence, so when I heard that, it gave me pause.)

P.S. I literally had to stop myself once I reached seventy minutes of music as a standard audio CD can only take 74 minutes of uncompressed stereo digital audio or about 650 MB of data, and with the .wav mix at 752 MB, if you plan to burn a CD, use the 160 MB .mp3 mix instead, or the individual .mp3 tracks.