Frequently Asked Questions

A selection of frequently asked questions.
Choose solution type
Choose product type

Yes, our Premium captions are compatible with our entire suite of encoders. This includes: 

  • Alta: A virtualized, API-powered IP video encoder that saves time and reduces costs with next-gen workflows.
  • Encoder Pro: An HD-SDI encoder trusted by major TV networks worldwide, offering low-cost, low-latency caption display.
  • Encoder 4K: A native 12G encoder that allows you to flexibly work with broadcast, live stream and in-room spaces while encoding in 4K or HD-SDI. 
  • Falcon: A budget-friendly cloud encoder that makes it easy to add captions to RTMP/S streams.  

Our SDI and IP captioning encoders are suitable for broadcast TV in the USA (in CEA-708 Standard), or in DVB regions like the UK, Australia, and the EU (in OP-47/DVB Teletext standard.)

All our IP and SDI caption encoders are compatible with third-party automated and human captioning services. However, by using AI-Media’s LEXI automatic captions or human captions, you can build an end-to-end solution without the costly and time-consuming need to integrate third-party offerings.   

No, the beauty of Ai-Live is that it doesn’t require any hardware. Both your speaker and audience can access Ai-Live via their web browser.  

Regulatory guidelines for live captioning vary greatly between countries and regions. While LEXI meets regulatory requirements for quality and time delay in many regions, some countries have specific requirements that LEXI does not yet meet. If you are unsure of your country’s requirements, please get in touch for more information. 

Generated by our expert human captioners, our Premium captions achieve up to 99.5% accuracy.While our LEXI automatic captioning solutions delivers results that rival human captions, at a fraction of the cost. LEXI captions deliver an average accuracy of 98.7%. 

Our LEXI automatic captions leverage the latest AI to deliver an average accuracy of 98.7% – which rivals the quality achieved by human captions, but at a fraction of the cost.  

Generated by our expert human captioners, our Premium captions achieve over 99.5% accuracy. 

Ai-Live can be paired with AI-Media’s LEXI automatic captions or human captions. Your speaker simply connects their microphone to their device and with one click, audio is connected. AI-Media’s captions are then displayed to attendees – either in person or virtual – via the accessible, browser-based Ai-Live caption viewer. 

Unlike most vendors, AI-Media is a one-stop shop of live caption solutions and tailored human services. We offer all the captioning technology you need in one place, including a range of caption delivery solutions that seamlessly integrate with our AI-powered LEXI automatic captions and Premium human captioning service. 

With a truly end-to-end solution from AI-Media, you can ensure you have all the technology and human services you need to caption your content, whether live or recorded. This eliminates the time-consuming and costly need to integrate third-party technology. 

Yes, LEXI automatic captions work with our entire suite of encoders. This includes: 

  • Alta: A virtualized, API-powered IP video encoder that saves time and reduces costs with next-gen workflows.
  • Encode Pro: An HD-SDI encoder trusted by major TV networks worldwide, offering low-cost, low-latency caption display.
  • Encode 4K: A native 12G encoder that allows you to fLEXIbly work with broadcast, live stream and in-room spaces while encoding in 4K or HD-SDI. 
  • Falcon: A budget-friendly cloud encoder that makes it easy to add captions to RTMP/S streams.  

The iCap Cloud Network offers a range of benefits, including: 

  • Higher bandwidth audio to captioners for increased clarity and word accuracy 
  • Enhanced security, with all iCap connections encrypted and authenticated 
  • Seamless transitioning from hardware-driven to software-controlled environments 
  • Seamless interoperability with AI-Media’s range of captioning solutions 
  • 24/7/365 connectivity to thousands of certified third-party caption partners worldwide 

Topic models (sometimes also known as custom dictionaries) are databases of specific terms and phrases that teach an Automatic Speech Recognition (ASR) engine, so it produces the words correctly when it ‘hears’ them. With LEXI, you have the option to add your own topic models. This process makes the live captions more accurate when consumers receive them to their screens. 

We offer a comprehensive range of on-premises, virtualized and cloud encoders covering SDI, IP, RTMP/S and 4K. Our most popular encoders include: 

  • Alta: A virtualized, API-powered IP video encoder that saves time and reduces costs with next-gen workflows.
  • Encode Pro: An HD-SDI encoder trusted by major TV networks worldwide, offering low-cost, low-latency caption display.
  • Encode 4K: A native 12G encoder that allows you to flexiIbly work with broadcast, live stream and in-room spaces while encoding in 4K or HD-SDI. 
  • Falcon: A budget-friendly cloud encoder that makes it easy to add captions to RTMP/S streams.  

Falcon is designed for time-limited streaming events like conferences or corporate town halls. It’s compatible with all major live streaming platforms, including Facebook Live, YouTube Live Events, Twitch, UStream, and many more. To use Falcon, you must be able to output an RTMP or RTMPS stream from a streaming media encoder.

Alta is best for higher-end IP video workflows on-prem or in public cloud, such as 24/7 linear broadcasts or an OTT platform. It can arrive pre-installed on a custom physical rack-mount server with broadcast-style front-panel controls, or as a virtual machine image to be hosted by the customer in a VM environment.

The CCMatch delay parameter allows our IP caption encoder users to match video and audio with live captions with perfect synchronization to the end viewer. CCMatch also aids in production and clipping by making sure live captions are contained completely within a segment, without overhang from programming into commercial blocks.

Alta is suitable for broadcast TV in the USA (in CEA-708 Standard), or in DVB regions like the UK, Australia, and the EU (in OP-47/DVB Teletext standard.)

For assistance please refer to the Manual: Captioner Guide: Using iCap with ProCAT software

For assistance please refer to the Product Manual: Captioner Guide: Using iCap with Rapid Caption

What are the key differences and relationship between SCTE-104 and SCTE-35?

SCTE-104 messages can exist either in VANC space of baseband (SDI) video, or sent between systems through TCP/IP, and are typically used as a precursor to eventual creation of SCTE-35 messages.

SCTE-35 messages exist inside an MPEG Transport Stream, and contain PCR timing data and possibly component PID data specific to that transport stream.

One way to think of this is that SCTE-104 messages are requests from an operator or automation system to create splice markers in a video stream, while SCTE-35 messages are the fulfillment of these requests, including all of the data that a downstream system will need to take appropriate action during live reception of the compressed stream.

The downstream system reading the SCTE-35 messages may be a server-side ad splicer, a set-top box, a video switcher, or any other equipment that can smoothly splice video from multiple sources into a single stream in the compressed domain. Based on the unique identifiers and flags in the SCTE-35 messages, the final operation may be insertion of an advertisement or other piece of interstitial video, overwriting of upstream video content with a locally generated slate, or segmentation of an in-progress recording into programs and chapters.

Which EEG products can insert SCTE-104 messages in VANC? Where do the messages come from?

EEG caption encoder models including HD490, HD491, HD492HD1490, and HD1452 are capable of inserting SCTE-104 messages into SDI VANC, with purchase of an add-on license.

The encoders have configurable built-in presets for a set of simple SCTE-104  “Splice Request” messages, which can be triggered with a GPI connection or from the encoder web configuration page. A selectable repeat interval allows continuous insertion of blackout triggers and similar repetitive content markers.

The EEG SCTE-104 encoders can also inject any SCTE-104 message received over TCP/IP port 5167 into the VANC space, either immediately upon receipt, or with use of an additional GPI firing trigger to support more deterministic frame synchronous injection. The TCP/IP communication method can be used for arbitrarily complex SCTE-104 messages including time signal messages with avail and segmentation descriptors.

The API for connecting to the EEG SCTE-104 encoders is described fully in the SCTE-104 standard. Use of the SCTE-104 automation-system-to-injector messages such as “Init Request” and “Alive Request” is encouraged, but can be omitted for many applications. The simplest application would involve automation or triggering software that simply opens a TCP connection to port 5167 on the encoder, sends a single packed SCTE-104 message, and then closes the connection when desired. The SCTE-104 message will appear in the VANC output of the encoder so long as it is a well-formatted multi-operation message structure according to the SCTE-104 standard.

EEG’s SCTE-104 messaging has been shown to be compatible with commercially available systems from vendors including Imagine Communications ADC, Snell Advanced Media, Motorola, Florical Systems, and DNF Controls, as well as a variety of proprietary customer systems.

Please contact us with other application development questions – we may be able to help prospective customers with sample SCTE-104 generation code in a variety of programming languages, or consulting on customized shims connecting existing production databases/processes to generation of SCTE-104 messages to support new applications.

Which EEG products can inject SCTE-35 messages into an MPEG Transport Stream?

Along with live closed caption encoding capabality, the Alta system contains a full SCTE-35 injection system for placing triggers into the incoming transport stream (IP unicast or multicast).

The triggers can be requested by any system using the SCTE-104 API over TCP/IP, through a configurable port on the iCap Alta instance.

Alta supports the full SCTE-104 “Simple Profile” – this profile includes Splice Requests, Splice Null Requests, and Time Signal Requests, with any number of DTMF Descriptors, Avail Descriptors, Segmentation Descriptors, and other proprietary/opaque descriptors. Alta does not currently support “component-mode” splices, encrypted triggers, or the schedule mode messages, but may in the future.

Please refer to the notes in the previous question about SCTE-104 insertion for more information about compatibility between EEG and other vendor devices, and the SCTE-104 TCP/IP API in general.

What PIDs does Alta insert SCTE-35 data on?

Alta selects a single unused PID based on the pre-existing PMT of the incoming transport stream, and modifies the outgoing PMTs to mark the inclusion of SCTE-35 data. The “cue stream type” descriptor is set to allow all SCTE-35 commands within this single PID.

Since only a single PID is assigned for SCTE-35, the “DPI PID index” field in SCTE-104 messages does not currently affect the placement of Alta’s SCTE-35 message output.

Which EEG products can monitor or receive existing SCTE-104 data in VANC?

The DE285, DE291, and DE1285,  SDI decoders show on-screen overlays of recent SCTE-104 splice requests, as well as providing a full XML-based log of all SCTE-104 triggers from the incoming video, with data visualization for each decoded field in the message. The EEG 1450 OpenGear card is also capable of triggering GPI outputs based on recognition of input triggers.

What is iCap Archives?

iCap Archives is available to encoder owners or caption service providers with iCap Broadcast Plus accounts. iCap Archives enables users with the correct permissions to retrieve a caption file for as-run data sent live over iCap between any pair of timestamps up to one month back from the date of the original transmission.

iCap Archives is popular for many applications including re-purposing live caption data for VOD distribution or future live re-airs, auditing on-air caption data for correctness, and troubleshooting discrepancy reports about past sessions.

What file formats are supported?

The native format is SCC. Other formats, including WebVTT, TTML, and plain-text transcripts, can be retreived by using the free conversion tool available with the EEG Cloud account you will create.

What are the prerequisites to use iCap Archives?

  1. The encoder must be on an iCap Broadcast Plus connection plan, as only these servers are capable of storing the data
  2. You must open a (free) account with your email address on to access iCap Archives
  3. If you plan to access archives by encoder name, you must have the iCap company and user name of the encoder, and an iCap Admin user name and password for the encoder owner company account
  4. If you plan to access archives by captioner name, you must have the iCap company and user name of the captioner, and an iCap Admin user name and password for the caption company user account

How do I get started using iCap Archives?

  1. Log into your account on (or register if you don’t have one)
  2. On the home page after login, go to “Our Services” and “iCap Archive Assembler”
  3. Fill out the request form with the required user account and time information. The fields marked “Company” and “User” refer to the encoder or captioner you are seeking records for. The fields marked “iCap Admin Username” and “iCap Admin Password” require your iCap Admin login for the SAME company as the encoder or captioner. This is the same set of credentials you would use to administer the company iCap account at If you do not have never used this site or are unsure if you have credentials, contact your company’s iCap administrator or  .
  4. When you click “Assemble Data”, you should receive an SCC file to download. Note that there may be no data to download if the encoder or captioner selected was not active during the time frame selected, or if the encoder was not on a Broadcast Plus server.

What time zone should I enter into the calendar input to download data?

You should use the time of the desired event in the local time zone set on your computer when downloading the file. It does not matter whether the caption encoder hardware is located in the same time zone as you or not, because all records are stored internally as UTC.

How do I convert from the SCC to another file format?

  1. Navigate to
  2. Use the “Select a File” control to choose the .scc file downloaded by your browser in the previous step.
  3. Specify an output format and click the Convert button.

Your browser will download the captions with the file format you specified.

Click “Download” to see a video illustrating these steps

What is the idle bandwidth for an encoder not currently in use but polling for new captioners joining?

Less than 5 kb/s.

What is the bandwidth for an encoder transmitting one channel of standard iCap encrypted and compressed audio data (or a captioner or monitor receiving the data)?

About 100 kb/s.

What is the bandwidth for transmitting or receiving standard iCap Video data (available as part of the Broadcast Plus subscription package)?

Video adds about 500 kb/s. This is at the standard rate of HD491 and HD492 encoders, which send 1-5 frames per second. The frames are designed to clearly show on-screen graphics, but action will generally not look smooth at this frame rate. The compromise is intended to preserve reliable low-latency performance even on limited network connections. Note that the caption monitoring overlay seen by the transcriber is produced locally at up to 30 frames per second, and will scroll smoothly despite the limited video frame rate.

What is the latency of the audio and video monitoring received over iCap?

The audio latency for remote iCap transcribers will generally fall in a range from about 250 milliseconds to 500 milliseconds. Precise times depend on the listener’s local buffer settings, and also may vary slightly with the ping time between the captioner, encoder source, and iCap reflection server.

iCap audio is sent over UDP with limited buffering and retransmission options. The iCap audio latency will never stack up over the course of a program as it would with some other technologies. However, iCap may be more sensitive than some types of less real-time applications to short bursts of congestion or high latency on a network.

Is iCap Audio latency affected by use of iCap Video?

No, as long there is sufficient bandwidth to support the video service at all, there is no direct connection between the video and audio delay. The streams are encoded, transmitted, buffered, and rendered separately.

Will increased bandwidth improve iCap audio performance?

iCap uses a relatively small amount of bandwidth per stream, but needs to have access to that bandwidth very reliably in order to see optimal performance.

When customers have a problem with iCap audio quality, most commonly the issue is with short-term bouts of packet latency or poor connection quality related to other software or devices temporarily congesting the network. Many common applications, especially ones that transfer large files, have a “bursty” usage profile, and this can cause temporary and unpredictable problems for more real-time applications like iCap.

In a larger managed network, implementing a form of QoS for traffic from iCap encoders may be a good idea to ensure optimal service.

iCap connects you to the largest network of caption service providers in the world. Click “Download” to view our list of iCap partner caption agencies.


You must first determine the method of communication between the iCap software and the writer software.

This will depend upon whether your writer software is running on the same PC as iCap, in which case the two pieces of software are communicating using a software serial port, or your writer software is running on a different PC than iCap, in which case communication is using a hardware serial port. Details regarding initial configuration is available in the manual, I will not go into further detail on this here. At this point, the configuration of the iCap serial port and the profile saved in the writer software should be correct.

Here is the procedure for using the “Test Serial Port” window:
STEP 1: From a clean system start, meaning that neither the iCap software or the writer software is currently running, launch the iCap software.
STEP 2: In the iCap client software, from the “Window” menu, select “Always On Top”. This will keep the iCap software visible even after you open your captioning software.
STEP 3: In the iCap client software, from the “Tools” menu, select “Test Serial Port” which will open the “Test Serial Port Settings” window.
STEP 4: Launch your writer software. Select the correct profile for iCap.
STEP 5: Begin writing. If the serial ports are configured correctly you should see the raw caption data in the “Test Serial Port” window (some of which may look like black
rectangles since some caption data cannot be represented by printing characters). If you do not see any data in the “Test Serial Port” window, refer to page 10 of the Troubleshooting section of the attached manual. Also verify your settings using Section 3, page 11 – Configuring Your Captioning Software to Use iCap.

Here is the procedure for logging into the EEG test encoder:
STEP 1: From a clean system start, meaning that neither the iCap software or the captioning software is currently running, launch the iCap software.
STEP 2: In the iCap client software, from the “Window” menu, select “Always On Top”.
This will keep the iCap software visible even after you open your captioning software.
STEP 3: In the “Access Code” box in iCap, enter “eegtest” (without the quotes and lower case) and click the “Connect” button. If you hear program audio and see captions in the monitor window then you know that you are connected to the encoder. If you enter an invalid Access Code you will see “Access Not Found” in the “Server Status” window. Reenter the Access Code and click “Connect”. If you are connected to the encoder you should see the message “SERVER (TIME) User Name has joined captionsolutions eegtest” in the chat window. The “Server Status” window should display “Connected”.
STEP 4: Click on the “Start” button to begin captioning. You should see the message “SERVER (TIME) User Name is captioning”.
STEP 5: Launch your captioning software. Select the correct profile for iCap.
STEP 6: Begin writing. In order to differentiate between the upstream captions already existing on the show and your captions, you can change the case of your data to contrast
the existing data. For example, if the existing captions are all uppercase, you can change your case to lower. This way when you see the captions in the iCap window you will
know that you are the source of the data.
STEP 7: When you have completed testing, send the blank and pass command from your writer software as usual to turn upstream captions back on, then hit the “Stop” button.

Need help with iCap for transcribing?


Access the Product Manual HERE

Before you start this tutorial, you may want to review:

1. Network Activated Licensing Requirements

One licensing option for many EEG products is floating licenses which are activated and validated through an internet connection. This includes CCPlay FilePro, Scribe, and Alta.

The computers running the licensed software must be able to reach out over HTTPS (port 443) to the domain “”. On any computer where typical web browsing is functional, no special licensing setup will be required. However, if you are allowing this outbound connection only through a special firewall exception, please add to your whitelist. If the whitelist only supports fixed IP addresses, or does not update DNS frequently enough for correct functionality, you can instead use (

For installations where no internet access at all is available, alternative licensing schemes such as hardware dongles for desktop software, or custom site licenses for other products may be available instead.

2. Alta Virtual Machine Set Up


OK, Now let’s begin:

You will need to have your license key string issued by your EEG Sales or Support representative.

Installing Your License Key

Once you have the virtual machine booting in your hypervisor (VirtualBox or other), perform the following steps:

1. Browse to the home page of your alta instance (for example

2. Click the License Key tab in the top level navigation

3. Enter your license key into the text input and click Save.

Now when you start a new Alta instance, the instance should begin as expected and the detailed logs will show a correct ability to license the software.

If the logs state that the licensing activation point could not be reached, you may need to modify the firewall/network access settings on your virtual machine controller, the host system, and/or the gateway router allowing traffic to pass out of your local network to the Internet.

See below for a video tutorial demonstrating these steps.

Download HERE

What is the default network configuration for a new Alta VM?

Usually you will need to explicitly enable one or more physical adapters to be assigned to the Alta VM in your VM hypervisor software (such as VirtualBox, VSphere, etc.). Most of these systems have several options for how the network ports will behave – for example in VirtualBox, “Bridged Adapter” puts the guest VM directly on the network the host is on, while “NAT” puts the guest VM on an internal network to its host, where it will not be directly accessible from outside the host. Please consult the documentation on your hypervisor if you have questions.

All interfaces in the Alta guest VM will initially be configured for DHCP.

How can I set static IP addresses on the Alta guest VM?

To set a static IP address, start the Alta VM, log into its desktop, and go to “Applications Menu” > “Settings” > “Network Connections”. Choose the appropriate NIC (which physical connections come up is determined through the VirtualBox menu described in the previous step – if there is any question about which NIC is which go to “Ethernet” > “Device MAC Address”  and match these values up to the ones in the VirtualBox menu), and go to “IPv4 Settings”. Change from “Automatic (DHCP)” to “Manual” if needed, and then add the desired IP address, netmask, gateway, and DNS servers as appropriate. For this work, the IP address and subnet have to be appropriate for the network that is being reached through the physical adapter that is reached through the VirtualBox settings for this virtual NIC.

Linux command line tools such as the “ip” command may also be used inside the Alta VM. Check the appropriate man pages or reference guides.

How many networks can Alta connect to simultaneously? How can multiple networks be used?

There is no specific limit set by EEG on how many networks an Alta VM can be connected to. You may be restricted by the number of physical networking ports on your host machine, or by the number supported by your virtualization software per VM.

It is possible to run all Alta-related functions (receive & transmit of media streams, web-based management, outbound connections to iCap/Lexi, and licensing validation) through a single network interface (NIC). It is also possible to divide these functions for reasons including increased bandwidth for media streams, QoS, and security separation of outbound iCap traffic and media networks.

The Alta VM will have a separate IP address for each separate network it is connected to.

The network for outbound iCap traffic should usually go through eth0 or the “first” network adapter if possible. This will prevent you from having to re-configure a different adapter as the “default route” for outbound traffic out of your local subnets.

For multicast media streams, use the “Multicast Interface” setting on each stream to set the IP address that for media transmit/receive.

For unicast media stream reception, provide the IP address of the interface to receive the stream on in the “input” field of the Alta channel. If only a port number is provided, unicast streams will be received across all interfaces on that port (and you must not send multiple unicast streams on the same port, even to different interfaces). For unicast stream transmission, the lowest numbered interface with a subnet matching the destination will be used (unless a different default route is set up in the Alta underlying Linux system).

For assistance please refer to the manual: Send & Receive SRT Streams and Captions using iCap Alta

For assistance please refer to the Manual: Using iCap with Advantage Software Eclipse AccuCap

Access the manual for EEG Web Application Security Policy HERE

For any questions regarding product setup, please reach out by phone by calling 516-293-7472, menu option #4, or via email at .

You can sign up for demos via your cloud services account, or reach out to any member of the Sales Team who can provide you with some brief paperwork, and once that paperwork is done, a member of the team will create any requested demo subscriptions.

Shipping costs are calculated only on hardware shipments. Shipping costs are based on the shipping address provided with the order, and can vary depending on customer request (shipping speed, Fedex or UPS, etc). A valid shipping account number (Fedex, DHL) is required from the customer for all international shipments. EEG does not invoice for international shipping.

Any sales taxes (if applicable) are based on the shipping address for any orders. Local sales tax rates/laws apply.

There’s a ‘Forgot Password’ button in the upper right-hand corner on the Cloud Services login page. You can use that tool to reset your password.

Currently, the Technical Support Team is reachable 8am-5pm ET Monday through Friday by phone or email. Overnight support is available 5pm-9am, as well as weekend coverage.

LEXI Local, iCap Local, LEXI Live and CaptionCast all need an EEG encoder connection. iCap Voice requires iCap PC.

Ready to Talk? 

Get in touch with our friendly team to learn how you can drive sustainable growth by partnering with AI-Media.

Start a conversation

Step 1 of 4
Which of the following best describes your business?
Please select one from the following options
Technical Support
Need technical assistance? Access technical support, product updates and upgrades, warranty details and more.
Go to the Support page

Become a partner

We’re collecting your details so we can respond to your query and send you relevant content. You can unsubscribe at any time by using the link at the bottom of our emails or writing to . In our Privacy Policy, you can learn more about how we handle your personal information, including about your rights and how to make a complaint.

Download Request Form