Smart glasses - Converts information into speech

Municipality of The Hague

Smart glasses that use artificial intelligence (AI) to convert images into spoken text to help visually impaired employees.

Last change on 28th of November 2025, at 7:49 (CET) | Publication Standard 1.0

• Publication category: Other algorithms
• Impact assessment: Field not filled in.
• Status: In use

General information

Theme

Work

Begin date

2019-09

Contact information

[e-mail]

Responsible use

Goal and impact

The purpose of this algorithm is to help visually impaired employees in the Social Work and Newly Sheltered target group with their work. These artificial intelligence (AI) glasses convert images into spoken language. This allows users of the glasses to work better independently. For example, it is used for reading labels, recognising things or finding directions at work.

Considerations

The glasses ensure that employees need less help and can work more independently. It also contributes to inclusion in the workplace. One drawback is that the glasses' voice recognition and image recognition do not always work well. This can cause confusion. The glasses are only an aid and do not make decisions themselves. Since a supervisor is always watching, using the glasses is safe and responsible.

Human intervention

The employee uses the information the glasses convert into spoken text to do the job. The glasses can mistranslate instructions, leading to errors.

Risk management

The risks are small because there is always supervision on the shop floor. It checks whether tasks are performed properly and makes adjustments if necessary. The user and supervisor also regularly discuss experiences with the glasses.

Legal basis

Participation Act

Operations

Data

Scanning smart glasses of work instructions and product information compiled by the work preparation department for the execution of work. In addition, environmental information such as room layout (such as corridor and office layout), signage and of predefined and appointed persons.

Technical design

The glasses scan text and use artificial intelligence (AI) to convert it into spoken language. To edit or translate the text, the user issues a command. When scanning rooms or the environment, the glasses describe the layout, signage and the number of people. The glasses can also give instructions, such as which way the user should walk. People can recognise the glasses only if a recording has been made of that person beforehand and the user has given that recording a name. In all these situations, AI does not learn on its own.

External provider

https://www.letsenvision.com/glasses/home