https://research.mach1.tech/ Recent content on Mach1 Research Hugo -- gohugo.io en-us ©2022 Mach1, Corp. https://research.mach1.tech/posts/microphone-arrays-and-vector-based-soundfields/ Tue, 12 Mar 2024 15:10:05 -0400 https://research.mach1.tech/posts/microphone-arrays-and-vector-based-soundfields/ Introduction This post is an account of multi-mic arrays commonly used for ambisonics and what was discovered when encoding them into a vector based or channel-based multichannel configuration such as Mach1 Spatial (14ch or 8ch). Challenges Capsule Layout Description Finding, discovering or measuring the capsule arrangement and ordering precisely since most of the time we will be skipping over the manufacturers conversion software as they tend to only target ambisonic encoding/decoding. https://research.mach1.tech/posts/naming-conventions_ambisonic-formats/ Tue, 28 Nov 2023 10:10:05 -0400 https://research.mach1.tech/posts/naming-conventions_ambisonic-formats/ Introduction This post explores publically described ambisonic nomenclature and dives into the terms proposed meaning and afterwards we share our opinion on these terms and how they may be more harmful than helpful to helping the community at large understand spatial audio. Within the ambisonic nomenclature there is scattered usage of various terms named [n]-format; they tend to try to describe nuances and differences in how the multichannel arrangement describes a fully spherical soundfield; however in our opinion we find that often these terms describe non-ambisonic concepts and create an issue where many users cannot differentiate ambisonic spherical harmonic content and more innate channel-bed or multichannel content. https://research.mach1.tech/posts/state-of-affairs-spatial-audio/ Fri, 18 Mar 2022 10:10:05 -0400 https://research.mach1.tech/posts/state-of-affairs-spatial-audio/ Over the last year there has been a lot of growth and interest into concepts surrounding spatial and multichannel audio in various different use cases and examples across many markets. We wanted to collect our observations and criticisms into a single post in an effort to help everyone researching to learn from our experiences and encourage better practices at large. When it comes to spatial audio we want to encourage growth as a medium and let different use cases define what spatial audio means in each instance while still catering to a creative production pipeline that enables more creative and technical growth at large, in human words: we want less proprietary approaches or tools and more open or cross-compatible growth. https://research.mach1.tech/posts/submitting-spatial-music-apps-to-ios-app-store/ Mon, 22 Nov 2021 10:10:05 -0400 https://research.mach1.tech/posts/submitting-spatial-music-apps-to-ios-app-store/ Status Minimum Features for Single Artist Music App: TBD Play controls and common audio player UI/UX Additional media (not confirmed) Fan interaction features Documentation to the App Review Board explaining your use of “spatial audio” Minimum Features for Music Distribution Platform (multiple artists): Not Confirmed Play controls and common audio player UI/UX Introduction At the time of writing this post “spatial audio” seems to be a term that has many varying meanings in the market. https://research.mach1.tech/posts/spatial-audio-defined-by-different-companies/ Wed, 20 Oct 2021 03:10:05 -0400 https://research.mach1.tech/posts/spatial-audio-defined-by-different-companies/ What do companies mean when they talk about spatial audio? Currently this is a very broad question because spatial audio is a subject spanning across disciplines and industries. We could talk about different flavors of audio spatialization, or about the history of spatial audio in academia. What we would like to focus on is the way the consumer electronics industry has most commonly employed the term in recent years, and share some points of clarification based on our years of experience both creating spatial audio experiences and the tools necessary to create them. https://research.mach1.tech/tutorials/transcoding-dolby-atmos-from-mach1-spatial/ Wed, 20 Oct 2021 03:10:00 -0400 https://research.mach1.tech/tutorials/transcoding-dolby-atmos-from-mach1-spatial/ To help encourage more agnostic mixing options for all kinds of multichannel and spatial audio formats and not encourage users to start their mixing process in a limiting format such as Dolby Atmos. We have been working on the best practices for conversion to/from Dolby Atmos and ADM metadata based formaats. This tutorial is to help as a starting point for anyone to test out converting to Dolby Atmos, the tutorial will focus on conversion from Mach1 Spatial, however you can safely convert to the Mach1 Spatial vector container easily from any other format transparently via the commandline tools and Mach1 Spatial SDK & Mach1 Spatial System in various ways. https://research.mach1.tech/posts/comparing-binaural-playback-ii/ Fri, 01 Oct 2021 03:10:05 -0400 https://research.mach1.tech/posts/comparing-binaural-playback-ii/ Introduction This post is a continuation of an exploration that prompted the development of an agnostic and process free multichannel/spatial audio framework that became the Mach1 Spatial SDK. In an effort to help showcase the sonic differences and usability of the current spatial audio market we are rendering various different format tests to sonically compare them without use of additional hardware, making A/B comparisons a little more accessible. It is recommended to create your own local tests and compare mixes played back in different means with headtracking active to further expose limitations. https://research.mach1.tech/posts/observations-and-limitations-of-apple-spatial-playback-implementation/ Wed, 22 Sep 2021 10:10:00 -0400 https://research.mach1.tech/posts/observations-and-limitations-of-apple-spatial-playback-implementation/ Introduction This post is to discuss our internal observations and limitations we found when transcoding to a ingestible 5.1 mix format and container that Apple will apply dynamic headtracking spatial playback for. We have done these tests from a source Mach1 Spatial file and used M1-Transcoder.app to correctly setup a 5.1 file, we have also done the same tests from the Dolby Atmos pipeline and can conclude that the same resulting file could be made from either workflow approach, knowing this we find it much easier to use anything and correctly transcode as needed via M1-Transcoder. https://research.mach1.tech/posts/observations-and-limitations-of-dolby-atmos-for-spatial-mixing/ Tue, 24 Aug 2021 10:10:00 -0400 https://research.mach1.tech/posts/observations-and-limitations-of-dolby-atmos-for-spatial-mixing/ Introduction This post is to discuss our internal observations and limitations we found when using Dolby Atmos for a creative spatial audio project. Using Dolby Atmos for front facing projected style content should be discussed separately from this post, we want to only focus on pain points we found when using and comparing Dolby Atmos for full spatial audio mixing and post-production when every direction is important from a first person perspective, not just the front. https://research.mach1.tech/tutorials/previewing-spatial-audio-on-apple/ Thu, 12 Aug 2021 03:10:00 -0400 https://research.mach1.tech/tutorials/previewing-spatial-audio-on-apple/ While investigating how Apple applies their own HRTF and headtracking implementation on top of Dolby Atmos it became appearent that there are a few glaring issues and limitations. It also became clear that we can transcode all formats to playback on Apple devices without use of Dolby Atmos at all, making a much cleaner and direct pipeline if your target is only for headtracking as defined by Apple. Apple seems to simply expect a video file with an EAC3 encoded audio track configured as 5. https://research.mach1.tech/posts/multichannel-codec-support/ Fri, 09 Jul 2021 03:57:53 -0400 https://research.mach1.tech/posts/multichannel-codec-support/ While working with spatial and multichannel audio across different target platform; what codec is available is not always what is preferred. This post is to keep tabs on the different expectations of each codec and how it handles multichannel audio streams/files. FFmpeg The following are FFmpeg’s handling of the codecs below. libopus (OPUS) We highly recommend using -c:a libopus instead of -c:a opus for proper multichannel handling from FFmpeg due to the following reason: https://research.mach1.tech/posts/feedback-on-apple-spatial-audio/ Mon, 08 Mar 2021 03:10:00 -0400 https://research.mach1.tech/posts/feedback-on-apple-spatial-audio/ The Mach1 Spatial team is adamant about sharing tools and info to help the community build mediums revolving around spatial audio and headtracking audio. We wanted to share some feedback we had regarding Apple’s approaches and what the next steps could be for them. We have been tracking and collating tools for various headtracking and IMU enabled devices as we find them and summarizing them here. This includes examples we make to evaluate headtracking hardware (and used for this post) such as: https://github. https://research.mach1.tech/tutorials/ffmpeg-useful-commands/ Mon, 20 Jul 2020 05:19:00 -0400 https://research.mach1.tech/tutorials/ffmpeg-useful-commands/ Flags Some common audio/video related flags or arguments to FFmpeg commands and what they mean. General: -i = input Audio/Video: NOTE: a || v are interchangeable -an or -vn = remove either audio or video -b:a or -b:v = bitrate, great for reducing a HQ video to work on mobile (example: -b:v 20m) -r = frame rate (video) -c:a or -c:v = refers to a command relating to the codec, so if you want to copy one of those codecs you can write: -c:v copy which is passthrough for video https://research.mach1.tech/posts/describing-3d-motion/ Mon, 15 Jun 2020 05:19:00 -0400 https://research.mach1.tech/posts/describing-3d-motion/ Describing 3D Motions & Rotations In the years of developing and creating for immersive mediums, there are usually moments of confusion and trouble when discussing and describing rotations or motions 3D (or 2D) space. Depending on what the project entails can lead to the creator describing things from 1st or 3rd perspectives and depending on what platform it is being built upon (and what frameworks) we see complications between different rotation and translation standards. https://research.mach1.tech/tutorials/custom-formats-with-mach1transcode/ Mon, 15 Jun 2020 03:15:37 -0400 https://research.mach1.tech/tutorials/custom-formats-with-mach1transcode/ The Mach1Transcode API and CommandLine executable m1-transcode can be supplied with a description of a custom multichannel format/configuration to be used as either an input desciption of a soundfield or an output soundfield. Developed originally to explore mixes on custom multichannel configurations often seen in various LBE or multimedia projects, however we found a lot of use cases for exploring custom pipelines between normally blackbox multichannel media handlers and serving them as needed to any other format/configuration. https://research.mach1.tech/posts/imu-enabled-devices/ Fri, 05 Jun 2020 03:10:05 -0400 https://research.mach1.tech/posts/imu-enabled-devices/ Introduction As more and more devices become available on the market for headtracking and positional tracking we can expect more oppurtunity for growth of spatial audio and other creative uses for multichannel/spatial audio. Here we will try to track and list known devices, focused more on the consumer side as the DIY side could be endless (however we will still list some here for anyone starting to research DIY solutions). https://research.mach1.tech/musings/an-introduction-to-the-history-of-spatial-music/ Fri, 28 Feb 2020 05:19:00 -0400 https://research.mach1.tech/musings/an-introduction-to-the-history-of-spatial-music/ Written by Visnja Krzic and Jacqueline Bosnjak An Introduction to Spatial Practices and Aesthetics Spatial practice in music originated in the early psalmodic practices of the Office and Mass. Antiphony reached its first peak in the concertato motets of the seventeenth century. During the eighteenth and nineteenth centuries, antiphonal practices were abandoned because the concern for balanced forces and blended sonorities became prominent. In the twentieth century, spatialization resurfaced again as a critical feature of sonic design and regained its compositional applicability and vitality. https://research.mach1.tech/posts/compression-applied-to-soundfields/ Thu, 05 Sep 2019 03:11:53 -0400 https://research.mach1.tech/posts/compression-applied-to-soundfields/ Targets of Testing To determine the value of compression in terms of file reduction compared to the quality of retaining the compression across the whole soundfield evenly. Also with discussion of the comparison of typical value of file reduction compared to the quality of resulting decompressed audio files. Also exploring if VBR compression types are safe to use on channel dependent soundfield material. Codecs Tested OPUS via libopus VORBIS via libvorbis MP3 via LAME AAC via FFMPEG Native AAC Codecs To Test AAC via FDKAAC AAC via apple Test Material Test 1: Wideband noise material in 90 degree offset combined into a soundfield with a narrowband saw wave @ 2kHz in the front direction and a sine tone panned 90 degree to the left. https://research.mach1.tech/posts/multichannel-order-from-codec-to-codec/ Fri, 15 Jun 2018 05:19:00 -0400 https://research.mach1.tech/posts/multichannel-order-from-codec-to-codec/ Introduction Another common pitfall when dealing with multichannel content or developing handling multichannel content is various hardcoded channel orderings that happen from leftover surround formats development focused around broadcasting. This now plagues more modern developments with various issues when attempting to bridge between solutions or development media pipelines specific to your project or a new application. Various commonly used media handlers know will read a multichannel of x channels and assume a specific legacy multichannel surround format, depending on the codec during handling the channels may be tagged, re-ordered and even filtered. https://research.mach1.tech/posts/comparing-binaural-decoders/ Mon, 05 Jun 2017 03:10:05 -0400 https://research.mach1.tech/posts/comparing-binaural-decoders/ Introduction In this post we will decode implemented spatial audio decoders and associated binaural filters and other HRTF effect processes. At the time of this testing we are selecting available ambisonic players and plugins as well as the Mach1 Spatial VVBP approach. The following are comparison of audio encoded into the following formats and decoded back to stereo in an effort to hear and compare audio processed in different spatial audio formats. https://research.mach1.tech/glossary/general-terms/ Mon, 01 Jan 0001 00:00:00 +0000 https://research.mach1.tech/glossary/general-terms/ Spatial Audio The term spatial audio has been thrown around a lot lately but what is spatial audio? What does it mean when you’re listening to something with spatial audio? Spatial audio is a concept that has been used and defined in several ways in our current audio landscape and marketplace. Some say that any audio expressed with more than one channel is spatial audio. The advent of stereo technology or surround sound systems can be considered spatial, but are not spatial audio as we define it today. https://research.mach1.tech/glossary/mixing-terms/ Mon, 01 Jan 0001 00:00:00 +0000 https://research.mach1.tech/glossary/mixing-terms/ Common Workflow Terms Panning Refers to panning a source sound or track into a spatial soundfield multichannel configuration or format. It is completely synonymous with Encode Synonyms : Pan, Encode, Input to Spatial Soundfield Related Terms : Monitoring Refers to the binaural downmix based on the current listening “orientation” as a method of monitoring or live listening to a spatial soundfield via stereo binuaralized headpones or speakers. It is completely synonymous with Decode https://research.mach1.tech/glossary/technical-terms/ Mon, 01 Jan 0001 00:00:00 +0000 https://research.mach1.tech/glossary/technical-terms/ Degrees of Freedom (DOF) 3DOF Typically refers to a single sensor (accelerometer or gyroscope) 3 axis output values Synonyms : Related Terms : 3DOF (Movement) 6DOF Typically refers to the usage of two sensors (accelerometer and/or gyroscope and/or magnometer) 6 axis output values Synonyms : Related Terms : 6DOF (Movement) 9DOF Typically refers to the usage of three sensors (accelerometer, gyroscope and magnometer) 9 axis output values Synonyms : Related Terms : 10DOF (and greater) Typically refers to the usage and combination of sensors for 10 or more axis data output values https://research.mach1.tech/glossary/existing-formats/ Mon, 01 Jan 0001 00:00:00 +0000 https://research.mach1.tech/glossary/existing-formats/ Deliverable Spatial Audio Formats Categories The breakdown of this page is to list some common components of spatial audio. Currently the major categories include: codecs & containers, formats & configurations and lastly standards & other components or tools. A codec is software that compresses and contains media data typically for transportation or storage as a file or files. Codecs are in a different category to formats typically, a ~good~ analogy to explain the difference would be that a codec is the box and shipping method for your packaging which is the format and the contents of the package is your spatial audio mix. https://research.mach1.tech/glossary/misc-terms/ Mon, 01 Jan 0001 00:00:00 +0000 https://research.mach1.tech/glossary/misc-terms/ Loose Terms Dimensions What does it mean when we say that an audio experience is 1D, 2D, or 3D? The interchangeable use of word ‘dimensions’ to refer to different things can be confusing, especially when the concept being referred to usually has its own name. One way we’ve heard ‘‘dimensions’’ expressed is to illustrate the difference between mono, stereo, and spatialised or binauralised audio, where 1D refers to mono, 2D refers to stereo, and 3D refers to spatial audio.