The Voiceitt app allows users to create a personalized profile by training the app with at least 200 phrases, capturing the user’s unique voice and speech patterns. This customization enhances the app’s ability to accurately interpret and translate the user’s speech, allowing for accessible communication. The user profile serves as the foundation for the communication experience within the Voiceitt app.
Voiceitt is designed to assist all individuals with speech and motor impairments who have difficulty communicating verbally. Some of those diagnoses include Cerebral Palsy (CP), Downs Syndrome, autism spectrum disorders (ASD), Parkinson’s disease (PD), Amyotrophic lateral sclerosis (ALS), Multiple sclerosis (MS), Muscular dystrophy (MD), Huntington’s disease (HD), deaf or hard of hearing and more.
Yes, you will need continuous internet access for Voiceitt2 to function.
The Voiceitt web app is primarily designed to cater to individuals with mild to moderate speech impairment. Voiceitt is most suitable for individuals whose speech is impaired but still mostly comprehensible by close family or caregivers. Individuals whose speech is severe and who are not generally understood by close family or caregivers are not good candidates for Voiceitt at this time. If individuals can be understood by their immediate circle on a regular basis, the web app can be considered a suitable solution for their communication needs.
English is the only language that is currently available and supported by the Voiceitt web application. However, we have plans to expand Voiceitt’s linguistic capabilities to include other languages in the future.
Yes, each child under the age of 13 is required to have a signed parental consent form to complete the Voiceitt account creation process. Currently parents are asked to print, sign, scan, and email the from to firstname.lastname@example.org.
Voiceitt has secure servers, firewalls, antivirus, SSL encryption and frequent database backups. Employees sign non-disclosure agreements and only authorized employees are provided restricted access to sensitive data. User data is deidentified and services utilize a user number to identify users uniquely.
No, the Voiceitt iOS application and Voiceitt2 are two separate products and do not communicate with each other. You will need to do training recordings separately to access each of these technologies.
The Voiceitt web app costs either $49.99 per month or $499.99/year. There is a 30-day free trial period before your credit card is charged.
If you subscribe to the Voiceitt2 service there is a 30-day free trial. You can cancel your subscription at any time during that 30-day period without any cost to you.
Yes. Call customer support at 800-729-0083, and you will be put in touch with someone at RAZ Mobility who can help you purchase multiple licenses for your organization.
A school purchases several licenses and obtains parental consent for each child under the age of 13 to give the students access to begin training in the Voiceitt web application.
After you subscribe to Voiceitt2, you will receive an email from RAZ Mobility with a link to the Voiceitt2 account creation page.
Yes, the Voiceitt CS team can change user email addresses upon request by the user.
Voiceitt can be used by anyone regardless of age, as long as they are verbal and have enough vocabulary to communicate wants and needs.
No, reading is not a requirement for using the Web app. During training, users can click the listen button to read each phrase aloud for the user to repeat and record. In these instances, having an assistant help with training can be helpful.
Once your account has been created, you will start the training process of recording 80short phrases. Recordings can be done over time as each session is saved. Phrases can be skipped for any reason by clicking on the “next” button on the right side of the screen. Once the training phase has been completed, you will receive an email that your speech model has been created and the web app can be used for speech recognition.
We have different training levels to accommodate everyone’s needs. Our standard training set is 3-6 words in length, our medium training set is 2-4 words in length, and our simplest training set is single words. Simpler training sets are suitable for users who have difficulty repeating the phrase as they are written, often skipping syllables or words.
We advise that individuals use the phrase set that best represents their speech and language ability. Individuals with robust language ability, speech that is produced easily, and who have no difficulty with our standard training set should continue to train as usual. Individuals who may not have a large vocabulary, skip syllables or words, or have speech that is produced at a slower rate should move to the medium training set. Individuals who generally speak in single words and have cognitive impairments should train using the single-word training set. Training sets can be changed throughout the process and are not static.
Only the Voiceitt CS team can change the user’s phrase set. Individuals or institutions do not have the ability to change datasets, however, eventually, this process will become automated, and, users will be given the option to switch their training set.
Speak Mode: Use this mode when you want to communicate with other people. The app will write your words on the screen and repeat what you have said in an automated voice. In speak mode, the app will stop listening automatically when there is silence or after a certain amount of time has passed. You should also use this mode when you want to communicate with ChatGPT.
Dictate Mode: Use this mode to transcribe long phrases into text for emails, notes, documents, etc. In dictate mode, the app continuously listens until you stop it yourself (click the blue ‘tap to stop’ button). The automated sound output is off in this mode. Dictate mode allows you to save notes and remove, copy, share, and edit your text.
If you say the name of a punctuation sign such as comma, period, question mark, exclamation point, etc… Voiceitt will put the sign you named.
For example, if you say: “Hello Exclamation point my name is Yubi period”
Voiceitt will dictate: “Hello! My name is Yubi.”
To communicate with ChatGPT, toggle to “speak” mode, press the ChatGPT button, then press “tap to speak.” Proceed to ask ChatGPT a question or give it a command. It takes a few seconds to generate a response.
You can say a single word or short phrase, and Voiceitt will output a longer phrase or sentence. To program a shortcut phrase toggle to “record,” click “shortcut phrases,” then click “add shortcut phrase.” Type in what you intend to say in the “when you say” text box, then type in what you want Voiceitt to say in the “Voiceitt will say” text box. Click “add,” then proceed to use your shortcut phrase in “speak” mode.
Personal Vocabulary allows you to add new words and phrases to your speech model vocabulary. This would be useful for adding personal names, locations, or a word that is consistently misrecognized. To program your vocabulary, toggle to “record” mode, click “personal vocabulary,” then click “add personal phrase.” Type in the desired word/phrase, then records it once by clicking the orange “tap to record” button. You may edit/remove a personal phrase in the future by clicking on it.
Please allow a few hours for your personal vocabulary to be incorporated into your speech model. Also, make sure the recordings of your personal phrases are clearly said. In general, to get your recognition up to par, continue to train general phrases.
Some of the keyboard shortcuts for record are:
Clicking space instead of tapping to record
Clicking on control in order to listen to your speech
Some of the keyboard shortcuts for speak are:
Clicking on T to use ChatGPT
Clicking on Alt to turn Voice On/Off
Clicking on space to start and stop recording
Clicking on enter to play your speech
Clicking on shift to switch mode to dictate
Some of the keyboard shortcuts for dictate are:
Clicking on R to remove words
Clicking on C to copy the text
Clicking on H to share the text
Clicking on E to edit the text
Clicking on space to start and stop recording
Clicking on enter to play your speech
Clicking on shift to switch mode to speak
Check and make sure your volume is on and turned up on your device.
If the text does not appear on the screen after you have spoken, there may be several factors involved. First, ensure your microphone is enabled and connected (see above). Next, be sure to press the large blue “speak” button and wait for it to say “listening,” then begin to speak. Be sure to speak at your usual rate and pause between phrases to give the app time to load text onto the screen. If your voice volume is very low, you may use a wired microphone to enhance your voice.
Generally, when recognition is low, we advise you to do more training (in “record” mode) to improve your speech model.
While you cannot directly edit the transcription of your recordings in Voiceitt, there are certain options available for refining the accuracy. Generally, you can do more training to improve recognition. If a specific word consistently faces recognition issues, you can enhance its recognition by adding it to your personal vocabulary. This allows Voiceitt to better understand and accurately transcribe the word in future interactions. In dictate mode, you have the ability to edit or remove words as needed, but this does not serve to improve the accuracy of your model.
To resolve issues with the speech/dictate button, you must ensure that your internet connectivity is stable and reliable. Additionally, check for any router-related problems affecting the functionality and consider refreshing the page to resolve any temporary glitches.
Currently, the Voiceitt Web App does not offer seamless integration with smart home devices. However, users can employ a workaround by playing the output voice of the command on Voiceitt, such as “Alexa, play my favorites playlist,” and ensuring that their smart home device is positioned close enough to the speaker to project Voiceitt sound to capture and process the command. This method allows for indirect communication between Voiceitt and the smart home device, enabling users to control their devices using Voiceitt commands.
In Voiceitt Web App:
First Time Setup
If using the Chrome browser:
Connect Voiceitt to a Teams meeting
All participants in the Teams meeting:
How to set up Webex:
Follow these instructions: https://voiceitt.com/voiceitt-for-webex
Voiceitt utilizes Amazon Web Services (AWS) servers which are located in Virginia, and the company also operates data centers in Belgium, Denmark, Ireland, Netherlands, Portugal, Sweden, United Kingdom, and United States.
As of now the technology is very compute demanding. As such it draws a lot of energy when in use. While in theory, it can run on a battery-powered, it is not expected to last for extended periods of time.
Yes, since the solution is personalized, it already naturally handles accents. Voiceitt’s data collection, processing and pre training pipeline is targeted towards impaired speech, but could, with some effort, be adapted to target accented speech.
Get support by phone, chat, email or through video communication.
Our Support Team is available Monday - Saturday from 10AM to 7PM ET.
1934 Old Gallows Rd. Suite 350
Tysons Corner, VA 22182