The case for multichannel playback

Stereo works. Two speakers, left and right, creating phantom images between them. It's been the standard for more than 60 years. So why bother with more channels?

Because modern playback systems have more speakers. 

A soundbar has 5-11 drivers. A car has 8-30. A home theater has 7, 12, or more. A live venue has dozens. These speakers exist, the question is what to feed them.

Option 1: Create native multichannel content.

This is ideal but impractical for most content. Dolby Atmos mixing requires stems, time, and expertise. Back catalogs remain stereo. Live sources arrive stereo. Streaming music is overwhelmingly stereo. The content pipeline hasn't caught up with the hardware.

Option 2: Upmix stereo to fill the system.

This is where HSR operates. Take the spatial information already encoded in stereo : panning, width, depth — and distribute it intelligently across whatever speakers exist. No content changes required. No workflow disruption. The listener's stereo library becomes immersive automatically.

Upmixing vs. native spatial audio

Upmixing isn't a replacement for native spatial mixing — it's a complement. Native Atmos for flagship releases; HSR for the other 90% of content that would otherwise play as stereo through two speakers of a twelve-speaker system.

You can also use Upmixing as a part of Dolby Atmos production, for your entire audio or for stems. We already provide an upmixing service doing so.

Upmixing vs. spatial audio systems

Full spatial audio systems (L-ISA, d&b Soundscape, SPAT Revolution) offer precise object positioning, room simulation, and complex routing. They're powerful tools for productions designed around them.

But they require:

• Content designed for spatial playback — object tracks, position metadata
• Dedicated workflow — spatial mixing sessions, soundfield design
• Significant infrastructure — processing hardware, control software, trained operators
• Time — spatial productions take longer than stereo

For a touring show, a broadcast facility, an installation, or a consumer product, this infrastructure often doesn't exist. The content arrives as stereo. The deadline is short. The budget doesn't include a spatial mixing session.

HSR fills this gap:

A theatre can use HSR to spatialize stereo playback tracks today, then upgrade to full object-based mixing for the next production, with our upcoming 3D audio solution. A broadcast facility can upmix legacy archives while building capacity for native Atmos production. An automotive OEM can ship immersive audio now, without waiting for the music industry to deliver multichannel content.

HSR is the practical path: spatial results from stereo sources, with minimal friction.

 The audio industry has spent decades trying to convert stereo content into surround sound. The challenge seems simple: create more than two channels from a stereo input. But every traditional approach introduces compromises that make the result less than ideal.

Mid/Side decomposition — the most common technique — splits stereo into "what's common to both channels" (Mid) and "what's different" (Side). The Mid goes to center, the Side goes to surrounds. This made sense when processing power was limited and the goal was simply "put something in the rear speakers."

But M/S upmixing creates fundamental problems:

FFT-based upmixers attempt smarter separation by analyzing frequency content. They can identify "center-panned" material more accurately. But they introduce their own problems

Modern speaker systems deserve better. A 7.1.4 Dolby Atmos setup has 12 speakers. A custom installation might have 32 or more. Feeding these systems with primitive upmixing wastes their potential.

HSR takes a fundamentally different approach. Rather than decomposing the signal into arbitrary components, HSR analyzes the spatial information already encoded in the stereo field and reconstructs it across any number of output channels.

The key insight: a stereo recording contains real spatial information. When a sound is panned left, it's louder in the left channel. When it's centered, both channels are equal. When it contains reverb, the channels are partially decorrelated. This information is precise and recoverable — if you analyze it correctly.

Stage 1: Correlation Analysis

HSR continuously examines the relationship between left and right channels:

This analysis identifies:

• Discrete sources — content with clear positioning (vocals, instruments)
• Diffuse content — ambience, reverb, stereo effects
• Position distribution — where energy sits across the stereo field

Stage 2: Spatial Extraction

From the correlation analysis, HSR extracts spatial components. Not just "center" and "sides," but a continuous distribution of energy across the full stereo width.

Stage 3: Output Distribution

The extracted spatial components map to your specific speaker configuration:

The distribution respects the original spatial intent:

• A source panned 30% left stays positioned 30% left, distributed naturally across your front speakers
• Diffuse content spreads across all speakers appropriately
• The total acoustic energy remains constant (iso-energy processing)

Time-Domain Processing

The entire HSR algorithm operates in the time domain. No FFT, no block processing, no frequency-domain transforms.

This means:

• 5 samples latency at any sample rate
• Zero spectral artifacts — no pre-echo, no musical noise
• Minimal CPU — runs on embedded platforms
• Sample-accurate — adapts instantly to content changes

Spacelite is a standalone application built on HSR. Connect any stereo source, define your speakers, and hear spatial audio in minutes

See product

Ultra-low latency

5 samples. That's the complete processing delay, regardless of sample rate.

For comparison, FFT-based upmixers typically require 1024-4096 sample blocks — that's 21-85ms at 48kHz. HSR is 200-800x faster.

This latency is imperceptible by any human and negligible for any synchronization requirement. Live broadcast, real-time monitoring, lip-sync critical applications — HSR handles them all.

True spatial reconstruction

Unlike M/S processing that forces content into discrete buckets, HSR reconstructs the actual spatial field encoded in the recording.

Example: A singer panned 20% left of center

Example: Orchestra with wide stereo image

Any output configuration

HSR doesn't care how many speakers you have or where they're positioned.

You define your speaker positions. HSR distributes accordingly. Asymmetric arrays, unusual layouts, dome installations — all work.

Iso-energy processing

The total acoustic energy remains constant through the upmix. Content doesn't get louder when spread across more speakers — it simply fills your array while maintaining the original loudness.

This is critical for:

• Broadcast — loudness compliance maintained
• Live sound — no gain surprises
• Consumer products — consistent listening level

Per-channel control

Every output channel has adjustable parameters:

Meta-parameters (in Spacelite)

 Spacelite exposes these meta-parameters in an intuitive interface, plus a full routing matrix with per-channel control. Ideal for live sound operators who need quick access to spatial

  adjustments.

Learn more

HSR vs. FFT-based upmixers

HSR vs. Matrix upmixers (Dolby Surround, DTS Neo, etc.)