What is voice banking?

Many people lose their voice or speech altogether as a result of their medical condition and require augmentative and alternative communication (AAC) to communicate. AAC devices mainly use generic synthesised voices which bear no resemblance to the person’s own voice.

Voice banking is a process for creating a ‘personalised synthetic voice’ (PSV), a synthetic approximation of a person’s natural voice (Costello, 2016).

To create a PSV, a set-list of phrases needs to be recorded – often between 350 and 1600. The PSV can then be used on an AAC device to ‘say’ an infinite number of novel words, phrases and sentences.

Technology in this area is progressing rapidly and the process to complete the voice banking recordings is relatively short, (an hour or two), depending on:

  • the choice of voice banking software
  • how well a person is feeling
  • level of respiratory support; and
  • fatigue.
Voice banking suppliers make various claims about the realism of PSV recorded voices using their technology. As a result, it is feasible to believe that a natural voice can be preserved by voice banking, and that a PSV can be used in everyday conversation in a similar way to natural speech.

As with any communication via an AAC device, there will be a reduction in speed and efficiency of a conversational turn; and it is unable to dynamically alter tone to communicate happiness, anger, sadness, irony or sarcasm for example,  (Pullin & Hennig, 2015) and using a PSV will not change this aspect.

However, a PSV:

  • will be unique to the individual who recorded it
  • is usually easily identifiable to the person who made the recordings and their family and friends; and
  • is different to the generic voices on AAC devices that are likely to be used by many others.
Please note: the resources and list of providers throughout this guidance are provided for informational purposes only. No endorsement is expressed or implied, and while we make every effort to ensure our pages are up to date and relevant, we cannot take responsibility for pages maintained by external providers

What is message banking and how is it different to voice banking?

Message banking is a process of directly recording individual words, phrases, sentences or personally meaningful sounds into a programme, thereby capturing the voice of an individual person with their own unique inflection and intonation.

Though unlike voice banking, a PSV is not created with this method of recording.

People often wish to record whole personal phrases, such as ‘I love you’ with their own voice, or routine daily requests that are unlikely to change. Calls to pets, laughter or other gestural sounds can also be recorded in this way.

People also message bank in languages not available in voice banking software.

The individual message banking recordings can be added to an AAC device, though playback is limited to only what has been recorded — they cannot be used for generating novel messages by spelling (Costello, 2020). People can record as many phrases as they wish for use on an AAC device.

Message bank recordings and a PSV can normally be utilised on the same AAC device, enabling more options for communication.

Message banking and voice banking should be discussed together and are complementary. Given the short time it may take to complete both, message banking and voice banking are frequently completed in parallel.

Who can benefit from voice banking?

Anyone at risk of voice change or loss, with the ability to use high technology AAC, may benefit from both voice banking and message banking.

Persons living with Motor Neurone Disease (MND) (plwMND) are probably the largest consumer group for voice banking. Acapela and ModelTalker report that approximately 80% and 95% respectively, of all the PSV files they create are for plwMND (Cave & Bloch, 2020).

Voice and message banking may also be considered for people with other progressive neurological conditions, for example Progressive Supranuclear Palsy (PSP) or Multiple System Atrophy (MSA).

It is important to consider the person’s cognitive function when considering the use of voice banking in all conditions, as this may change over the course of their condition and impact on their ability to use high technology AAC.

People with head and neck conditions, including preparation for laryngectomy, long-term planned tracheostomy or glossectomy may also benefit from voice and message banking.

In addition, banked voices may be donated to AAC users who have never been able to speak, for example due to congenital conditions, in order that they can benefit from having a personalised voice that suits their age and regional accent.

Role of the speech and language therapist

Speech and language therapists (SLTs) have an important role to both educate and to provide practical support regarding voice banking and message banking.

SLTs do not need to be experts in AAC to fulfil this role, as support is available once the need to use AAC arises. You also do not need to be a ‘techy’ person – the level of IT knowledge we all use in our day to day work is sufficient to support the voice banking process.

Voice banking and message banking are best carried out before there is significant change to speech quality.

Patients should be provided with enough information as early as possible to make their own decision whether to proceed.  In order to fully support this, speech and language therapy services should accept referrals for patients who do not yet have any difficulty with their speech or voice and may not yet have a formal diagnosis of MND or another condition.

Early referral is beneficial for allowing discussions around potential future options for AAC, and the availability of professionals to advise and support people regarding AAC, has been shown to improve AAC device uptake (Baxter, Enderby, Judge, & Evans, 2012).

It is important to confirm, at this stage, whether the person is likely to choose to use high tech AAC if their speech/voice alters. If they are only likely to choose low tech options with no voice output, voice banking is not an appropriate intervention.

Early input from the SLT is also important for forming a good working relationship with a patient and their family. This will help with all aspects of a condition as it progresses. It may also offer opportunities for discussion around other key areas, such as advance decision making, while communication remains ‘easier’.

It is important that the SLT sets expectations around voice banking and message banking. It is also worth noting that the PSV can never be an exact replica of a person’s voice and therefore some people may be disappointed and feel that the voice does not sound like them.

If available, SLTs should demonstrate their own PSV downloaded to an AAC device to practically demonstrate:


  • what a PSV is
  • how it differs from natural voice; and
  • how it is used as part of an AAC system.

Many voice banking suppliers do not charge SLTs and other health and social care professionals to record their voice and download their PSV to AAC for demonstration purposes.

A discussion about options for voice banking and message banking can be sensitive. It may be difficult for patients to consider the possibility of a time when they may have no voice.

Some patients may not wish to consider voice banking while their natural voice is unaffected, and others may be frustrated that they had not been told early enough to be able to complete the process with their unaffected voice (Cave & Bloch, 2020).

Discussing voice banking and message banking in terms of ‘insurance’ –  in case voice may change in the future – is a helpful way to frame the conversation (Benson, 2015).

As ‘preservation of identity’ may be a major factor in deciding to voice bank (Cave & Bloch, 2020), the patient should be provided with information to make their own decision, even where their voice has already been affected by dysarthria.

Voice banking with dysarthria or dysphonia will result in a PSV with dysarthria or dysphonia – there will be no ‘correction’ of any impairment. Clinical experience indicates that this may still be acceptable and useful to people; conversely some people will record with unimpaired speech and voice yet not find the resulting PSV acceptable, and choose not to use it.

For those people with impaired speech or voice, the SLT could discuss using a ‘donor’ voice to create a PSV — possibly a friend or family member who sounds similar — as an alternative.

The SLT may also discuss digital legacy recording – the creation of video messages for family and friends. Digital legacy recording may be confused with voice banking (Cave & Bloch, 2020) and it is important to clarify that voice banking is for the purpose of use on an AAC device.  There are many ways to record a video message and Record me now (www.recordmenow.org) can be used as a framework for this.

We suggest that the SLT advises on the equipment needed, which may be a computer and a headset microphone, or a smartphone or tablet. Some services may loan equipment. The MND Association has an equipment loan service for plwMND (www.mndassociation.org).

As the voice banking process can often be completed in an hour or two, the SLT may train the patient to record independently or a significant other to support. They could provide speech and language therapy assistant support, or directly offer help themselves.

It is not essential for the SLT to be present for recordings. In some areas, the MND Association have trained Association Visitors in the process of voice and message banking, and may be able to provide assistance.

If the patient does eventually need to use AAC, the SLT can assist with downloading the voice onto the device. The assistive technology regional hubs may provide support for this if their AAC equipment is being provided.  Individual AAC software providers may also offer support with the download process into their software if needed.

There are costs associated with voice banking, but these are incurred almost always at the point of download to an AAC device. This means patients can normally record and create a PSV without delay or cost, however, funding will be needed to download the PSV to an AAC device if it becomes required.

The MND Association may provide funding for plwMND.

Impacts of voice banking

Clinical experience frequently reports positive feedback from patients who bank their voice, and from their significant others.

Choosing where and when to speak with a PSV is a highly personal decision and differs from person to person. Some choose to use their PSV only with specific people or situations, and then use a more generic synthetic voice at other times.

‘Preservation of identity’ is an overarching theme for plwMND deciding to voice bank or not (Cave & Bloch, 2020). Participants deciding to voice bank considered it would help to maintain their identity and preserve their social and work networks.

Participants deciding not to voice bank highlighted it could not replace their natural voice or preserve their identity.

However, few in either group showed an awareness of how a voice bank is used in AAC, and how communication using AAC is significantly different to natural speech (Cave & Bloch, 2020).

A PSV has been reported as saving the distress of significant others being unable to remember what their loved one sounded like, a kind of ‘vocal insurance’ in case voice is lost (Benson, 2015).

A PSV is argued to promote human interaction and social participation, enabling plwMND to function more effectively within their own families and social circles and allow them to contribute to wider society for longer through continued employment (Nathanson, 2016).

Significant others are usually supportive of their partner’s decision regarding voice banking and some felt a PSV could support their relationship as they try to manage the difficult times ahead (Cave & Bloch, 2020).

A PSV may be beneficial for AAC uptake as the behaviour and attitudes of communication partners towards AAC have been shown to  influence how and where AAC is used and ultimately how successful the provision of AAC may be (Scherer, Jutai, Fuhrer, Demers, & Deruyter, 2007).

Videoed interviews of two plwMND who were already using their banked voices (used for training of SLTs) indicated significant others may value a PSV more than plwMND themselves (Cave & Gleave, 2016).

However, no research has been identified that investigates the value of a PSV used in everyday communication (Linse, Aust, Joos, & Hermann, 2018). No quantitative data has been identified showing the difference between use of a PSV and a standard synthetic human voice available in many AAC devices  (Nathanson, 2016).

Jůzová reports that a significant improvement to quality of life was achieved by creating PSVs for several patients prior to laryngectomy. However, there is no indication of how they measured the change in quality of life or how the study could be replicated. Jůzová acknowledged that further research is required to show empirical evidence of any change in quality of life (Jůzová, Romportl, & Tihelka, 2015).

As voice banking is still a relatively new intervention, there remains a lack of evidence for the claimed benefits of it. However, there is significant and growing interest in voice banking, particularly within the MND community.

In 2018 the MND Association directly helped over 100 plwMND to voice bank. More than 600 registrations are forecast for 2020 (Cave & Bloch, 2020).

Practical next steps for SLTs

If you are an SLT interested in voice banking and think that it could benefit a service user, then we suggest these next steps:


Bank your own voice

Bank your own voice with at least one of the voice banking providers and download it onto an AAC device. This will make conversations much easier once you have something to demonstrate and can boost your confidence talking about the process. (At the time of writing) you may be able to bank your voice without charge from Acapela my-own-voice, Cereproc, ModelTalker and Voicekeeper. Other providers may help if requested.


Check your equipment

Check whether you have the right equipment – you may need a computer and a headset with a microphone, or just a smartphone, depending on which provider you choose. You will also need an AAC device with software for demonstration. The MND Association can loan equipment to help people living with MND to bank their voice – an online loan request form is on their website.


Consider your client group

Consider your client group and how you get referrals through for them. Is this early enough? Do you need to revisit your pathway and talk directly to referrers?


Consider funding as early as possible

Almost every voice banking supplier allows the necessary recordings to be completed prior to charging for their service. This is particularly helpful if your patient’s voice is beginning to deteriorate. However, funding needs to be considered as far ahead as possible too, as voice banking is not usually paid for by the Specialist AAC hubs.

You may need to apply for funding from the local Clinical Commissioning Group (CCG) or a charity and this will take time. For plwMND the MND Association may be able to fund part or all of the cost. The MSA Trust may also be able to fund part or all of the cost for people living with Multiple System Atrophy.


Consider how to ‘frame’ voice banking

Think about how to explain voice banking using the information you have read here. In our clinical experience, framing voice banking in terms of ‘insurance’ is very helpful.


Keep their options open

Do not exclude service users who already have dysarthria from the opportunity of making their own decision about voice banking. Set out the facts and options, including the possibility of another person who sounds similar donating their voice.


Work through the process

Find one patient who may wish to try voice banking and work through it together. If your patient has MND, the MND Association can sometimes offer remote support to you and your patient through the voice banking process – contact communicationaids@mndassociation.org for more information.


Consult your peers

Suppliers make various claims about their voice banking products though as with many commercial suppliers, if it sounds too good to be true it probably is.

It is helpful asking other professionals with voice banking experience what their recommendations are. @VoiceBankingMND on Twitter and Facebook is managed by richard.cave@nhs.net and followed by a large group of SLTs and other professionals supporting others to voicebank, and is a good forum to ask questions. Free voice banking and message banking training sessions are regularly arranged for SLT teams.


Involve significant ‘others’

It can be a difficult task for service users to think about the most important sentences they wish to record for the future. If you have a patient who wishes to message bank, ask if they wish for significant others, friends and work colleagues to become involved too. Clinical experience has shown patients are not always aware of their own speech ‘isms’, phrases or words they say all the time. Other individuals’ perspectives will help your patient create a bigger volume of phrases to store.

Embrace technology

Don’t be frightened to embrace the technology – this is a great tool in our SLT toolkit! Be ready to use it before a patient asks you – because someone will.


The following resources can be found on voice banking.

And please note: the resources and list of providers throughout this guidance are provided for informational purposes only. No endorsement is expressed or implied, and while we make every effort to ensure our pages are up to date and relevant, we cannot take responsibility for pages maintained by external providers.

Motor Neurone Disease Association resources:

General resources:

Case studies

Read the following case studies on patients who have experience of voice banking:

Useful contacts

Individual voice banking providers:

Other organisations

  • Parkinson’s UK
    A Parkinson’s research and support charity in the United Kingdom. Their aims are to improve the quality of life for people affected by Parkinson’s and find a cure for the condition
    Email: hello@parkinsons.org.uk
  • MSA Trust
    The MSA Trust is the UK’s support and information service for people with MSA, their families & carers. It also funds research to find the cause & cure.
    Email: support@msatrust.org.uk 
  • MS Trust
    Working to provide information for anyone affected by multiple sclerosis, education programmes for health professionals, funding for practical research and campaigns for specialist MS services.
    EMail: hello@mstrust.org.uk
  • Macmillan
    Are one of the largest British charities and provides specialist health care, information and financial support to people affected by cancer.
    Visit their website for contact details
  • The Sequal Trust
    A communication disability charity that aims to improve the quality of life for those living with severe learning disabilities, mobility issues and communication difficulties. It doesn’t matter if someone has incoherent speech or is completely non-verbal, when they become a member of The Sequal Trust, they will liaise with the person’s assigned SLT to ensure they provide the most suitable assistive technology such as an eye gaze, Grid Pad or Lightwriter.
    Phone: 01691 624 222
    Email: info@thesequaltrust.org.uk 


Baxter, S., Enderby, P., Judge, S., & Evans, P. (2012). Barriers and facilitators. International Journal of Language and Communication Disorders, 47(2), 115–129. https://doi.org/10.1111/j.1460-6984.2011.00090.x

Benson, J. (2015). Have your MND patients taken out their vocal insurance yet? Bulletin, (December), 12–13. Retrieved from http://www.callscotland.org.uk/common-assets/cm-files/files/voice-banking-for-patients-with-motor-neurone-disease.pdf

Cave, R., & Bloch, S. (2020). Voice banking for people living with motor neurone disease: views and expectations (manuscript in preparation).

Cave, R., & Gleave, R. (2016). Dr David Turner interviewed about Voice Banking by Richard Cave of the RHN [Video File]. UK. Retrieved from

Costello, J. M. (n.d.). ALS Augmentative Communication Program | Message BankingTM | Boston Children’s Hospital. Retrieved May 1, 2020, from http://www.childrenshospital.org/ALSmessagebanking#

Costello, J. M. (2016). Message Banking vs. Voice Banking: A Very Successful Proactive Model for People with ALS/MND. In 14th Annual Allied Professionals Forum. Dublin. Retrieved from https://www.alsmndalliance.org/wp-content/uploads/2016/12/costello-message-banking-handout.pdf

Jůzová, M., Romportl, J., & Tihelka, D. (2015). Speech corpus preparation for voice banking of laryngectomised patients. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 9302, pp. 282–290). https://doi.org/10.1007/978-3-319-24033-6_32

Linse, K., Aust, E., Joos, M., & Hermann, A. (2018). Communication matters-pitfalls and promise of hightech communication devices in palliative care of severely physically disabled patients with amyotrophic lateral sclerosis. Frontiers in Neurology, 9(JUL), 1–18. https://doi.org/10.3389/fneur.2018.00603

Nathanson, E. (2016). Native voice , self-concept and the moral case for personalized voice technology technology. Disability and Rehabilitation, 8288(May), 73–81. https://doi.org/10.3109/09638288.2016.1139193

Pullin, G., & Hennig, S. (2015). 17 Ways to Say Yes: Toward Nuanced Tone of Voice in AAC and Speech Technology. Augmentative and Alternative Communication, 31(2), 170–180. https://doi.org/10.3109/07434618.2015.1037930

Scherer, M., Jutai, J., Fuhrer, M., Demers, L., & Deruyter, F. (2007). A framework for modelling the selection of assistive technology devices (ATDs). Disability and Rehabilitation: Assistive Technology, 2(1), 1–8. https://doi.org/10.1080/17483100600845414


This guidance has been developed on behalf of the Royal College of Speech and Language Therapists (RCSLT) by:

  • Jen Benson
  • Rhiannon Evans
  • Richard Cave

With thanks also to the group of reviewers for their valuable feedback in the development of this guidance.

1  of  10