Apr 13, 2022 | Read time 3 min

Earth Day 2022: Voice Technology and Our Environmental Choices

This Earth Day, our CEO, Katy Wigdahl, discusses the environmental impact of machine learning and what we all must do to help our planet. 
Earth Day 2022: Voice Technology and Our Environmental Choices

Earth Day 2022: Voice Technology and Our Environmental Choices

On Friday 22 April 2022, we celebrate Earth Day. This year, I want to speak openly about the climate crisis. Specifically, I would like to take this opportunity to look at what we as a fast-growth voice technology business can do as a collective, as individuals, and how we can all work to emphasise the organization’s main theme this year, #InvestInOurPlanet.

As Earth Day’s website makes clear, “Unless businesses act now, climate change will ever more deeply damage economies, increase scarcity, drain profits and job prospects, and impact us all.” It’s on all of us to do something before it’s too late. So, what role do we in tech – and specifically voice tech – play?

Technology’s Environmental Footprint

At Speechmatics, we’re self-aware enough to know the great advantages our technology brings comes with consequences. In voice recognition, training huge models can lead to large emissions of greenhouse gas. While modern hardware is much more efficient than the hardware of the past, energy usage is still at a problematic level.

The complex technology behind CPUs and GPUs takes energy and material to produce, while scarce and raw materials are often polluting to mine and becoming harder to find, despite demand rising. In a broader sense, the technology sector’s environmental impact cannot be ignored. In 2019, the world produced roughly 53 million tonnes of electronic waste - that figure is projected to grow by 33% - it is now the world’s fastest-growing waste stream.

On Earth Day and beyond, we must all do our best not to ignore the sourcing and production of these materials and ask: what can we do about the growing problem surrounding the disposal of electronic waste linked to our technology?

What We’re Doing

We believe in doing what we can when we can. In the last year, Speechmatics has responsibly dealt with waste at a total collection weight of 993kg. In this period, we’ve recycled a total of 492 kilograms, giving a recycling percentage of 50%. Through an environmentally responsible waste handler, 0kg of waste we’ve produced has ended up in landfill, with estimated CO2 savings of 90kg through direct, indirect, and avoided emissions.

Additionally, we're looking at introducing an electric vehicle scheme to allow UK-based employees to purchase environmentally-friendly vehicles on a salary-sacrifice basis. This is on top of encouraging staff to use reusable items and limiting single-use plastics. To support this, each member of the team is given a reusable, insulated water bottle when they join us.

When Speechmatics employees travel, e-boarding passes are used, and non-stop flights are booked wherever possible. Airports are chosen with the location in mind being as close to homes and destinations as possible. We’re also looking into carbon offsetting our travel impacts, although we admit, we could be quicker with this.

Invest in the Future

We’re proud of what we do as a business, but we acknowledge we could do more. As for the technology industry as a whole, we can all see the impact we have and know that the scales are tipping perilously in the wrong direction. But we also know that great technology has the potential to get us out of this crisis. Investing in the planet means looking to the companies who are making a real, positive impact on how we live.

To repeat a core Earth Day message, “Smart companies are discovering that it is no longer a choice between going green and growing long-term profits — sustainability is the path to prosperity.”

Here at Speechmatics, we’re on the side of the smart companies.

Katy Wigdahl, CEO, Speechmatics

Latest Articles

Carousel slide image
Technical

How to build a microbatching workflow with the Speechmatics API

Build a cleaner path between batch and real time. Learn when micro-batching makes sense, how to chunk audio, submit jobs, stitch JSON, and scale safely with the Speechmatics API.

Speechmatics
SpeechmaticsEditorial Team
Carousel slide image
Product

Alphanumeric speech recognition: why voice assistants mangle SKUs (and how to fix it)

A guide for voice AI engineers, ecommerce platforms and warehouse teams on SKU recognition accuracy voice assistant deployments depend on: why speech recognition systems produce transcription errors on product codes, what to measure when error rates matter, and the fixes that move the needle on order picking, voice ordering and customer-facing voice AI.

Speechmatics
SpeechmaticsEditorial Team
Carousel slide image
Technical

The Adobe story: How we made cloud-grade AI work on your laptop

Behind the build: what it takes to make cloud-grade speech recognition work inside Adobe Premiere, and why Whisper raised the stakes.

Andrew Innes
Andrew InnesChief Architect
Carousel slide image
Company

Adobe and Speechmatics deliver cloud-grade speech recognition on-device for Premiere

Adobe Premiere users can run the most accurate on-device transcription locally; efficient enough for a laptop, powerful enough for professional work.

Speechmatics
SpeechmaticsEditorial Team
Carousel slide image
Use Cases

Best speech-to-text AI guide: APIs, platforms and services compared

Speech-to-text has moved from novelty to enterprise infrastructure. Here's how the leading platforms stack up in 2026 — and how to pick the right one.

Tom Young
Tom YoungDigital Specialist
Speechmatics x Thymia combine medical-grade speech-to-text with clinical-grade voice biomarker intelligence to identify health signals.
News

AI can now understand health signals from 15 seconds of your voice, including fatigue, stress and type 2 diabetes

The joint platform returns transcription and health signals in real time, with no additional hardware required.

Speechmatics
SpeechmaticsEditorial Team