Frequently Asked Questions

Welcome to our FAQ page! Here you’ll find answers to the most common questions about Speech Graphics technology, products and services. We’ve gathered everything you need in one place to help you get the answers you need quickly - but if you have a question that isn’t answered here, help is just a click away.

  • SGX is our award-winning, audio driven facial animation production software. Built on powerful AI and the unique Speech Graphics muscle-dynamic system, SGX is used widely across the video games and entertainment industries to deliver high-quality facial animation from audio alone and at massive scale via robust batch processing tools, while keeping creative control in the hands of animators. It produces accurate lip sync as well as nonverbal behaviors including full-face emotional expressions, head motion, breath, blinks and eye darts. See also What is SGX.

  • Animation production happens in a standalone system – SGX Production Tools. For character setup, pre-visualization, and animation import, the SGX Studio plugins for Maya, Unreal and (soon) 3dsMax are used. However you can also export animation in a plain-text format that can be imported into any engine without plugins. See Exporting.

  • No, Speech Graphics software can generate animation for any type of facial rig – whether driven by bones, blendshapes, or UI controls. We don’t prescribe any specific set of blendshapes, layout of bones, or control interface. But generally speaking, if your rig is capable of achieving FACS-like movements, then it will be well-suited to Speech Graphics.

  • Yes, a rig system like Epic’s MetaHumans works very well with SGX and produces great results.

  • SGX also automatically drives emotional expressions, head motion, blinking, eye darts and other non-verbal behaviour like efforts, grunts, breath and laughter.

  • Typically, the ratio of processing time to input duration is roughly 1:3. This means it takes about 1 second to convert 3 seconds of audio into animation. When not using transcripts, processing is faster.

  • Yes, with SGX Director you can quickly and easily make high-level edits to your animation and review changes immediately via our live-link feature. See SGX Interactive Processing.

  • Yes, our comprehensive knowledge base is available in English, Mandarin and Japanese.

  • Yes, we offer free evaluations so you can put the software through its paces. Just contact us for more information.

  • We have various SGX packages to suit a wide range of situations, whether you’re a small independent studio shopping for a publisher, an established AAA Studio, a publisher with multiple studios, or something in between.

    The cost is calculated based on the number of concurrent users you would like to cater for in your animation team, the language modules you might need and the number of character setups requiring support.

  • Yes, Rapport is the leading platform for delivering real time, interactive experiences with 3D avatars.

    Built on Speech Graphics’ award-winning, audio-driven animation technology, Rapport delivers expressive, full-body and facial animation for live, interactive experiences across learning, customer engagement, and immersive environments.

  • As SGX uses a universal model of human speech to map audio to articulation, it works well with any language - even fictional ones. 


    However, a number of specific language modules are available to help achieve the best quality when combining audio with text transcripts. These include English (with dialects for the UK – England/Wales and Scotland – and the US), French, German, Italian, Japanese, Korean, Mandarin, Polish, Portuguese, Russian and Spanish (European, Mexican). See also
    Language Support.

  • There is no minimum machine spec, but processing speeds, and the number of parallel processes you intend to run should be taken into consideration. Some basic recommendations for optimizing SGX for high-volume use include:

    • Use a dedicated server or desktop, rather than a laptop or similar.

    • Minimum 16GB RAM.

    • An Intel chip will have highest performance for our libraries given we use MKL.

    • In terms of processors, cache per core is more important than number of cores or clock speed.

  • SGX operates on a floating license. You will receive a license file that provides credentials to our cloud-based license server. Each of the products (SGX Studio, SGX Director and SGX Producer) has its own pool of allocated seats. Additional seats, products and languages can be issued by changes on the server side.

  • Please contact us with any further queries. Our team will be happy to hear from you.

  • You can download the appropriate software from our Downloads site. You will need to sign up using a registration code for your organization.

Our products set the standard in audio-driven animation:

  • SGX - Automate accurate lip sync and complete nonverbal behavior from audio alone.

  • Rapport - Create high impact, interactive avatar experiences for training, sales and marketing driven by AI

Proudly Working With:

Would you like to know more?

  • We work with AAA and indie games studios, filmmakers, and all kinds of artists alike.

  • Contact us today to discuss our products and packages to find the best solution for your project.