VoiceBankNOW is a 501c3 non-profit 100% volunteer organization, providing virtual voice education and
resources with impartiality. VBN is not affiliated, associated with or funded by any other company.
How do I Voice Virtually?
record
generate
your voice
The process of creating your personal Synthesized Voice starts by banking, or recording, your voice. You will need a quiet space, a microphone, and a computer/tablet. Chose which Traditional Voice Synthesis Companies you will record with, they will have different requirements, and you will record online with their service. For your AI Voice Cloning, you will create your own recordings. See 01 Traditional Voice Synthesis and 02 AI Voice Cloning below.​
Once you have recorded with or submitted your recording to a company, they will 'do their magic' and generate your custom voice. Depending on the service you chose, this could take anywhere from a few minutes, to a couple weeks. The resulting voice will vary depending on the quality of your input, the quantity of material and the company synthesizing or cloning. The results are personal, which again, is why we recommend recording with as many companies as you can.​
The cost also varies. While in most cases it will be free to record and hear a sample of your synthesized or cloned voice, the cost and terms to USE it varies.​
Your voice then becomes a 'plug in'. It is the information that software applications will need to generate speech that sounds like you!​​
app
device
speak!
You will chose a software application to 'plug' your voice into and use via Type to Speak, TTS. Some voices will pair with some applications, and not others. There are of course, pros and cons to different applications. See our guide to Voicing Apps.
The app will be installed on your device, a phone, tablet, or a dedicated AAC (alternative augmentative communication) device.
VOICE, APP, and DEVICE must all be compatible! Compatibility is shown in more detail in our Guides to 01 Traditional Voice Synthesis, 02 AI Voice Cloning and 03 Voicing App.
Through the magic of technology, you then can type or text to generate speech that sounds like YOU!!
01 Traditional Voice Synthesis.
A Traditional Synthesized Voice can yield a voice that sounds a lot like your natural voice. However, they tend to be a bit monotone. There is neither expression nor intonation. There is usually a one time fee to purchase your Synthesized Voice. In addition to using on phone or tablet, these work on the dedicated AAC hardware. So if you have mobility issues and will need larger buttons for simpler communication or eye tracking, you definitely will want Traditional Voice Synthesis.
​
For the vast majority of services, you will need to currently be able to speak, and can expect to record for 2 to 7 hours (not all at once!). Some voice synthesis companies can provide enhancements on your voice, so if its already a bit compromised this is still an option!
​We recommend you record with more than one company, if possible, because results vary per user as to which is your 'best voice'. Different synthesized voices are also compatible with different applications. ​​​
Even if you want to do AI Voice Cloning, we strongly recommend you still create a traditionally cloned voice. Cloning services charge monthly with character use limits, ie how much you speak. Software applications to use AI Voice Clones for daily communication are just coming to market. To date we only know of ONE, the Whisper app.
See 01 Traditional Voice Synthesis for information on companies that provide this service.
02 AI Cloned voice.
Cloned Voices can be indistinguishable from your natural voice. There have intonation, cadence, speed, and will interpret your tone of voice based on content. Developments are happening where you can chose 'expression' such as happy, surprised, fear, angry. Cloning companies typically charge a monthly ongoing fee to use the cloning services. Each time you generated voice it is generated 'in the cloud' engaging the cloning company, and thus you must have internet connection. We highly recommend preparing a recording for AI Voice Cloning, as this is rapidly going to replace the traditional voice synthesis.
​
AI Voice Cloning requires a high quality digital recording (.wav) that is a minimum 2 minutes to 2+ hours . The longer recording, the better the resulting cloned voice will be. One of the easiest ways to achieve a high quality recording is with a Zoom H1 essential digital recorder.
AI Cloned Voices are currently primarily limited to use on a tablet or computer. While they offer a realistic and rich voice, they have been developed mainly for use in the world of film, business applications, voice overs, and gaming. The processing is done online through the providers website. Internet speed will be a factor and you will wait as processing happens to convert your text into speech. Cloned voicing works fabulous to create presentations, or other pre-recorded messages. Companies charge monthly with per character limits and as such, may not be affordable for everyone. *vbn is working to change this*​​​​
While there are phone apps for Cloned Voices they are not designed as daily communication tools and thus are too clumsy and often expensive, to use conversationally. At current, the only communication based app we are aware of is the Whisper app (and its FREE!). This arena is rapidly evolving, so expect to see great things in the near future!​
See 02 AI Voice Cloning for additional information on recording your voice and companies providing Voice Cloning services.
HELP! I already lost my origin voice!
VBN is working toward getting information about voice synthesis in the hands of every person at risk, so they have the opportunity to Voice Bank NOW. If you have already lost your voice your options for synthesis are more limited, but you are not out of options! Traditional Voice Synthesis is available with pre-recorded voice clips from a few companies. And if you can find at least 2 minutes of good quality recording of your voice, than AI Voice Cloning might be right for you.
03 Use your VOICE!
Losing the ability to speak can be a deeply traumatic experience, often leading to feelings of isolation and a diminished sense of self. Regaining a part of that identity through a virtual representation of your own voice can be transformative. It fosters confidence and provides a more vibrant way to express yourself and connect with loved ones.
After your AI Cloned Voice or Traditional Synthesized Voice is created, you will need to 'plug it in' to a Voicing Application on your device. Since voicing apps can differ significantly, we have compiled a list of our top recommendations for Voicing Apps.
Utilizing a Voicing App on your smartphone offers great convenience. However, trying to text quickly to maintain the flow of conversation can be challenging. Typing on a tablet, which lacks the tactile feedback of a keyboard, may also slow you down. Therefore, it's worth considering a Bluetooth keyboard. Additionally, since phone and tablet volume can be limited, pairing with a Bluetooth speaker is advisable.
We understand that not everyone is quick at texting or familiar with touch typing. While frequent texting and typing can enhance speed over time, many apps offer customization options with 'buttons' for frequently used words or phrases. For those mobility challenges, an AAC device can serve as a helpful alternative.
Dedicated AAC (augmentative and alternative communication) devices provide a wide range of 'buttons' with pre-set phrases, which can be extremely beneficial for individuals with mobility challenges, along with eye tracker bars. These devices do require custom programming for setup, and while tech-savvy users may manage this, we highly recommend collaborating with a speech-language pathologist (SLP) or a manufacturer's representative. For more information on these systems, check out our Guide to AAC hardware. Please note that funding for AAC hardware can be a challenge, as many insurance companies do not cover them, and costs can range from $5,000 to $10,000.