Skip to main content

Voice Development

Voice Development

Scroll to content

Voice Development

Voice Development is part of every Conversational AI strategy

Good voice development starts with the understanding that voice is still a new technology in the Conversational AI market, especially in DACH. This includes all the technical components involved to build a robust architecture for clever and intuitive conversations.

Why Voice Development?

Development of voice components

In conversation design, voice UX, and linguistics, voice technology has also triggered a new understanding of the development of voice components and the various specialties.

Creative voice developers need to implement an excellent technical base: voice development involves a higher degree of strategy than you might think.

Right technical and linguistic tools

Every conversational application needs a strategic choice to pick the right technical and linguistic tools. There are many providers to select from to build stable recognition, choose the right APIs, provide databases, and implement a custom TTS.

With custom-build assistants, it’s essential to make good choices to guarantee an independent development process, robust data security, the right data analytics, and ultimately deliver a premium voice user experience.

Short-term solutions will not lead to success for any voice application.

Developing Custom-Built Assistants

At, we share the creative process between design and development in creating custom-built assistants with our client's development experts. – Voice Development – Infographic explaining ASR, NLU, TTS

In these voice projects, we work together with ASR, NLU, TTS, dialogue, voice backends, data analysis specialists, and many teams that share their knowledge and creativity to work on our shared vision of useful conversational AI. Every project, agile team constellation, and user interaction let us learn more about voice technology and Conversational AI.

We feel our shared knowledge in every team and community gives us the responsibility to design and implement sustainable, responsible and human-centered, and safe Conversational AI.

Custom-Built Experiences with Alexa and Google Assistant

As millions of other designers and developers do, we also love to use the familiar voice platforms' extraordinary technical framework and user experience.

When implementing voice applications for Amazon Alexa or Google Assistant with our clients, our linguists and developers make sure to make the best of development and design for implementation on those platforms. For example, in German, the most natural conversational experience for users with Amazon Alexa is the status of a 'native Skill' (or app), which is why we tackle this challenge with every voice application we design and implement.

Both platforms also provide an excellent framework for multimodality, accessibility, and a seamless device and user experience. It's often a great base to use this framework and increase the number of branded Skills for Alexa or Actions for Google Assistant.

This enables us to improve the experience through an exclusive sound UX and individual conversational branding. – Voice development – Amazon Alexa Voice Assitant, Echo show, placed on a table with person making tea and asking for the weather

Every conversation starts with understanding

Clever voice development always starts with the quality of speech and language recognition, especially in languages other than English.

Whether it’s a rule-based or machine learning approach, our voice developers and linguists make sure that users are understood in the language they speak – all while avoiding bias in AI.

Voice development often needs to find a creative solution to meet legal requirements, protect data, and ensure harmless interaction with AI. These requirements differ according to market, project, and user needs.
Dr. Laura Dreessen / Senior Voice User Interface Architect & Consultant at

Voice development needs stable implementation processes

Conversational app development needs a vision, just like other software development. In voice development, this means establishing roles and procedures to ensure a reliable and professional process.

The voice backend is the necessary foundation for intuitive conversational interaction: the more intelligent the implementation, the more natural the conversations. – Voice Development – infografic explaining the process from a stable process to voice development to a natural conversation

We keep user conversations private

Security and privacy are of utmost importance for our team, and we always comply with European regulations. We don't store users' voice recordings or transcripts: we are privacy keepers.

Good conversations need a long-term strategy

A voice project is never 'done' since conversations are never a static thing. Conversational AI learns and improves through user interactions. And this is why an overview of how your app performs and how users interact and speak with it is so important.

Usually, this overview includes changing or adding a voice recording, improving a language model's recognition, updating unclear prompt and dialogue flow hiccups, or choosing new features based on analytics. This part of the voice development lets you formulate and validate a hypothesis and measure your best option in making the best voice UI you can. – Voice Development – long term strategy; icon with mountain

There is no voice development without design

Voice development directly influences and empowers the conversation design process through technology, constantly perfecting it and quickly pointing out its limitations. Voice designers and voice developers need to act as a team and continuously challenge technology to create digital assistants that show conversational skills – and thus good usability and UX.

Although anyone can develop a voice application for the big platforms, it’s the expert collaboration of voice developers, linguists, and designers that makes for excellent voice applications. – Voice Development – Photo of Amazon Alexa and Google Speakder

Five things I struggle with when trying to explain NLU to managers

  • #AI NLP
  • #VUI Design
Discover blog article