ACE Flex 2

The ACE-FLEX-2 is a server based automatic speech recognition system that uses a cutting-edge computational linguistics program to process speech into text and caption data from audio content for live captioning purposes.

When two channels of audio are needed, the ACE-FLEX-2 is the perfect solution for captioning content with high accuracy, low latency, and multiple output options.

The ACE-FLEX-2/ENC includes a built-in caption encoder, eliminating additional equipment and creating an easy all-in- one workflow.

Server Based Solution

The ACE-FLEX-2, like all ACE series products, is an on-premise solution that does not require an internet connection to function. Although an internet connection is not required, connecting to a network provides streamlined workflows and remote GUI access. The ACE-FLEX-2 can receive two separate audio inputs, it can de-embed the first two channels of audio from two SDI or HDMI inputs. Audio from a streaming source is another way that the system can receive audio. Each channel is run independently and can caption simultaneously with two RS-232 outputs to two caption encoders. Multiple customizations can be programmed into the units inserters including, speaker identification, number of characters per line, number of lines visible and more. Once a captioning task has been completed, a file is generated with selected transcript files. Some of the file types include text files, caption files, subtitle files, or time coded text files.

Intuitive GUI with Extensive Customization

The ACE-FLEX-2 system has a user-friendly web Graphical User interface (GUI) where you can perform all relevant caption processing and administrative functions. Caption processing includes, 24 hour scheduling, event creating, task creation, creation of speaker identification, caption translation, and more. Speaker identification processes voice prints of speakers to create identification tags used to denote individual speakers within the captions. This unit includes in unit translation from English to Spanish/Spanish to English without an internet connection. As an administrator, users can be created and set up with unique parameters, tailoring the use of the ACE-FLEX-2 to any environment, such as broadcast stations, university departments, school districts, and shared usage throughout a municipality.

The ACE series US English dictionary is made up of over 500,000 words with daily language model updates to the unit when connected to the internet. The language models are continuously updated by scanning websites and internal processing of audio during system set up. Users can also load unique vocabulary such as, names, cities and geographical locations. Beyond the US English dictionary, the ACE series supports the following languages: English (US, UK, Ireland, Canada, Carribean, Asia, India), Spanish (Spain, US-East, US-West, Puerto Rico, Latin America, Mexico, Peru, Argentina, Chile, Colombia), Catalan, French (France, Canada, Belgium, Switzerland), German, Italian (Italy, Switzerland), Swiss, Portuguese (European, African, Brazilian), Basque, and Dutch (Netherlands).