Aira vs Speakonia

Aira is a service that helps visually impaired individuals to live independently. In doing so, we — through the latest, personalized assistive technology and approaches — serve as visual interpreters and navigators for blind and low-vision people. Here is a detailed description of how the Aira platform works: Smart glass technology: The blind user puts on a Smart glass that has a camera, network connectivity and a rich set of sensors, such as Google Glass and taps on a button. An “Uber” style routing will connect the user to an available trained Agent to help the user with his or her request. Aira Agents are certified and trained: Agents are specially selected and trained through a 4-step process (described below). Agents are either from the user’s friends and family circle, or from our Aira Agent Team. The user will always interact with Agents whom they feel comfortable with and have trained/worked with in the past. Agent selection and training is a four-step process: — -Agents are selected by a pre-screening process that is similar to tests of a 911 operator. — -Every Agent is trained by experienced Aira team members on the use of the Agent Dashboard and go through various emergency protocols. — -The Agent is also introduced to “Aira Phraseology,” which will help him or her deliver timely, clear and concise information while taking into consideration the user’s surroundings. — -In addition, Agents go through a certification process after undergoing a thorough background check and signing confidentiality agreements. The Agent has a sophisticated Agent Dashboard which has the following data feeds: — -The current view from the user’s Google Glass User’s current location. — -Current traffic conditions, points of interest, government warnings, etc. — -Personal data gathered from the users — such as medical history, emergency contacts, things that they enjoy doing, places they like visiting, and names and contact information of family members and personal assistants. — -Facial recognition that will enable the user to identify individuals around him or her. Agent-User interaction: Agents are first paired with users to get acquainted with users’ needs by working with them on simpler tasks under supervision. This provides an opportunity for both the Agent and blind user to get to know each other and develop a compatible relationship. Once the user builds trust and becomes comfortable with the guidance received, the Agent will transition to guiding them independently with their tasks. User safety is paramount: — -Our platform, electronic equipment, system protocols, Agent training and user acclimatization are designed to put user safety as our top priority. All the transactions via Aira platform are secured confidential. — -The system has automatic algorithms to assess potential dangers. For example, the system is constantly monitoring for sound, video input, and sensor feed. As new safety systems and wearable devices become available, Aira’s platform is built to incorporate them and remain at the leading edge of safety. — -Aira has mechanisms to handle contingent situations, such as device or technical failure. Agents will be able to make calls on behalf of the user, request taxi service from reliable service providers to take the blind user home safely, or follow a specific user’s contingency plan. — -Even when network coverage is lost, Aira continues to store the user’s location locally on the wearable device. Once the network connection is regained, their location history is updated on the server, enabling the Agent to pinpoint the user’s current location. — -Aira’s platform is also capable of solving latency issues (slow network) for when there is limited network bandwidth, such as in crowded or remote areas. Environmental adaptation: In order to provide precise and accurate guidance in a noisy area, the majority of communications between the Aira Agent and the user can be done via pre-set tones and clicks. Latency adaptation (aka Slow 4G LTE Network): The Aira team has worked diligently to reduce time effects due to network delays. New algorithms being developed will optimize the quality of video streaming and reduce network delay to 1–2 seconds (depending on network reception in the area). Extensibility: — -Aira will be able to leverage indoor locations, based on systems such as iBeacon and Google Indoor Maps in the future when the technologies mature. — -Aira was built using its API platform, which will be open to external developers, allowing them to expand it to most of the industry’s wearable technologies. This means Aira, as a system, can incorporate information from other wearables, cameras, proximity sensors, infrared canes and other sensors to provide a complete solution for users. — -Aira has access to next generation Artificial Intelligence algorithms and advanced object and pattern recognition technologies. Market research and experiments: — -Aira has conducted experiments at FFB Dining in the Dark 2014, Vision Walk 2014, LA Tech talk, CSUN 2015 and Braille Institute. — -There have been 50+ blind user experiments, 200 blind user in-depth interviews, and 5 sessions for Agents and training. — -There are 100+ beta users signed up for trials in summer 2015. — -There are 8 formal focus groups with Mobility and Orientation specialists, assistive tech experts, and rehab advisors.

Regardless of their skill level, the typical user can use Speakonia very easily. In a technological sense, the graphical user interface is comparable to Notepad. The text-to-speech voice of their choice will read anything the user types into the box that resembles a play button once they click it. The user can switch the voice at any moment by selecting a different voice from the drop-down box under the Speech Properties that is titled Voice. With the sliders beneath the drop-down menu, the user can additionally alter the voice's volume and pace. Several languages can be put alongside it. Interface The menu bar, toolbar, text box, and "Speech Properties" Box are the four basic components of the Speakonia user interface. The Menu Bar is mostly used for storing and exporting files, as well as for software registration. It is not the program's primary focus. Shortcuts for saving features are available on the toolbar, along with buttons for starting, pausing, and stopping the voice reading of text that has been entered into the text box. You will spend the majority of your time in the software in the Text Box. To make the voice say something, just write it into the text box. Last but not least, Speakonia's most useful feature is the Speech Properties window. You may swiftly switch between voices, alter the voice's pitch and volume, as well as its speaking speed. The useful "Lexicon" feature in this window also enables you to amend a word's pronunciation if the voice pronounces it incorrectly. Speakonia's Speech Properties window. Voices of Microsoft Speakonia comes with at least 11 voices by default. Installing a second bundle of voices titled simply: Microsoft voices Mary, Sam, and Mike will significantly raise this number. The Microsoft voices as well as a variety of other voices will be available for you to use after the download is finished (RoboSoft 1-6, whispers, etc.). creating a.WAV export The spoken text can be exported from Speakonia as.WAV audio files. You need to register with Speakonia in order to export a file as a.WAV audio file. One only needs to provide their name and email address to register. Once enrolled, you will then have access to all of Speakonia's services after receiving a code through email. Anyone who wants to use the voices in their YouTube videos must register. Trivia The Lernout & Hauspie TruVoice voices were really modified to create Speakonia's default adult voices. They too have had new names. There are no markers for pitch, speed, or volume to show levels of each. A Registry Editor is the sole method. The Microsoft Voices that come with Windows XP have different voice patterns from those in Speakonia. Instead, it resembles Microsoft Sam for Windows 2000. Although Speakonia is no longer supported, CFS-Technologies still offers the program for download on its website. (Also accessible via CNET's download page) Windows RT is not compatible with Speakonia. Every TTS Voice's test voice starts with the words "This is a test for Speakonia." The following registration key is required in order to utilize Speakonia to its fullest extent: Name: Unrestricted Permit Email: [email protected] Reg. Identifier: SPK-174101-172083 References The Mediafire links might not function. Use the Google Drive links if it doesn't work. http://www.cfs-technologies.com/home/ is the download link. id=1.4 Pieboy6000 has supplied links to Mediafire. Speakonia: Where can I find MediaFire? 21142f5uq5u4ket Sam, Mike, Mary, and RoboSoft voices from Microsoft: http://www.mediafire.com? e4hkxgt5jbsxu72 The Speakonia website no longer offers the ability to download the additional voices, such as British voices. But you can download them from here thanks to Ducky. There are additional links for the voices (including more voices than were previously provided) here or here.