Whispp Voice AI at Interspeech 2025 | Fair and Inclusive Speech Science and Technology

Rotterdam Ahoy, 17–21 August 2025. Theme: “Fair and Inclusive Speech Science and Technology”.

We’re at Interspeech 2025 to raise awareness for assistive voice technology and what it can mean in everyday conversations. This is personal and practical: people with whispered or affected voices should be heard without effort, at work and at home. Progress is not a number on a slide. Progress is a conversation that feels easy on both sides.

On Saturday 16 August we sponsored the Young Female* Researchers in Speech Workshop at TU Delft. It was energising to see students and mentors connect. Thank you to everyone who took part, and a special thanks to Johannah for helping organise the day.

The first two conference days were all about listening and learning. Thanks to Karolina and Johannah for being on the ground with us. Your conversations sharpened our focus. We’re on site through 21 August; Joris and Akash are available to connect during the remaining days. If you want to talk, come by the booth or send us a message and we’ll find time between sessions.

Alongside universities and research labs, many well known companies are here. When teams from Apple, Google and Meta join the discussion, ideas move quickly. That makes this week a good moment to align on what “good” should mean for assistive voice technology.

What “good” should mean

Good means less strain for the listener and more confidence for the speaker. Fewer repeats. Natural timing that lets people jump in without talking over each other. Clear choices about privacy and control, spelled out in plain language. Above all, the user’s experience, not just a score, defines success.

How Whispp approaches it

After awareness comes the how. Whispp’s voice technology turns an affected or whispered voice into a clear, natural sounding voice while keeping words, timing and intent intact. It is designed for live, natural back and forth first, with calling available when it helps. The aim is not to change who you are; it is to let you speak as yourself and be understood.

People decide what is processed and when. We keep controls simple and explanations direct. If you prefer a voice that sounds like you, we support that. If you prefer a neutral voice, we support that too. The point is choice and comfort.

Why we’re writing this now

Awareness comes before adoption. Many people still don’t know assistive voice technology exists, or they assume it’s only for clinical settings. It isn’t. The conversations we’re having this week: in sessions, hallways and at the booth, help us show what’s already possible and hear what still gets in the way.

If you’re at Ahoy, let’s talk about outcome‑based evaluation, respectful data practices and small product decisions that reduce stigma instead of adding to it. If you’re following from elsewhere, we’ll share takeaways after the conference.

Thanks again to the YFRSW community for the strong start, and to everyone who has stopped by so far.

Interested in how Whispp works, but couldn’t make it to Interspeech 2025? Send us a message through our contact form.