Apr 4, 2018 | Read time 2 min

Speechmatics’ Custom Dictionary helps live broadcasters never get names wrong in subtitles again

 Media and Comms Monitoring_RGB_video with keyword trigger

The easy-to-use interface allows customers to quickly add context specific words to Speechmatics’ dictionary simply via text Today, Speechmatics is announcing the launch of Custom Dictionary, a new feature for the company’s current speech-to-text transcription offering. Live broadcast subtitling has a recognized problem when it comes to names and place names. Custom Dictionary (CD) lets users add context specific words in real-time such as footballer names or breaking news locations, instantly increasing the accuracy.

Traditionally, transcription providers would have to train each specific model for a customer to include special words. However, setting up a new model has typically been very time-consuming, usually requiring an overnight rebuild. Custom Dictionary streamlines the entire process by putting the customer in control and allowing them to add new words in plain text to “teach” the model, regardless of use case or language context. CD enables users to define the context of a conversation in advance, and input those difficult terms into the dictionary, so that when they do come up they are transcribed without error. For example, sports subtitling is dramatically improved with prior input of player names, venues and club names. While developing CD, accuracy remained a top priority, as customers indicated this to be key to improving their value propositions. Speechmatics has successfully improved on its accuracy rates and already comprehensive dictionary by dynamically adapting to the context of individual use cases. When compared directly to Google’s Cloud Speech, and with both Speechmatics’ CD and Google’s API “hints” features enabled, Speechmatics produced a transcript that was more than 12% more accurate*.

Benedikt von Thüngen, CEO at Speechmatics, explained:

“Accuracy is still the key metric for speech recognition systems and it has remained top of mind throughout the development of our solutions at Speechmatics. Understanding what has been said is key and everyone speaks using different acronyms and terminology that is bespoke to them. With CD we now offer the first step towards personalised transcription. At Speechmatics, we pride ourselves on unparalleled accuracy rates, and Custom Dictionary allows us to take our speech-to-text transcription from great to awesome. People often don’t realise how much language is evolving, it’s nowhere near as static as you’d think. This means that as a transcription provider, we have to keep up with this evolution and address our customers’ needs, regardless of their industry. Custom Dictionary is the perfect solution to this problem – not only is it incredibly simple to use, but it’s also very smart and can recognise context and pairs of words to make sure it gets the spelling right.”

Ian Firth, VP of Products at Speechmatics, explained:

“In case of the broadcast industry, for example, closed captioning is an area that desperately needs improvement. From the outdated re-speaking approach to very awkward blunders, the lack of innovation can cause a lot of errors. Custom Dictionary simplifies and improves closed captioning for broadcast.”

The technology can be used in a number of industries, from broadcast and surveillance services, to financial organisations and conduct risk assessment, a vast number of companies use transcription for keyword spotting, real-time captioning and compliance. To try a demo of Custom Dictionary at NAB in Las Vegas, 7th – 12th April, visit Speechmatics at booth SU7813.

Latest Articles

[alt: Bilingual medical model featuring terms related to various health conditions and medications in Arabic and English. Key terms include "Chronic kidney disease," "Heart attack," "Diabetes," and "Insulin," among others, displayed in an organized layout.]
Product

Speechmatics achieves a world first in bilingual Voice AI with new Arabic–English model

Sets a new accuracy bar for real-world code-switching: 35% fewer errors than the closest competitor.

Speechmatics
SpeechmaticsEditorial Team
[alt: Illuminated ancient mud-brick structures stand against a dusk sky, showcasing architectural details and textures. Palm trees are in the foreground, adding to the setting's ambiance. Visually captures a historic site in twilight.]
Product

Your voice agent speaks perfect Arabic. That's the problem.

Most voice AI models are trained on formal Arabic, but real conversations across the Middle East mix dialects and English in ways those systems aren’t built to handle.

Yahia Abaza
Yahia AbazaSenior Product Manger
new blog image header
Technical

How Nvidia Dominates the HuggingFace Leaderboards in This Key Metric

A technical deep-dive into Token Duration Transducers (TDT) — the frame-skipping architecture behind Nvidia's Parakeet models. Covers inference mechanics, training with forward-backward algorithm, and how TDT achieves up to 2.82x faster decoding than standard RNN-T.

Oliver Parish
Oliver Parish Machine Learning Engineer
[alt: Healthcare professionals in scrubs and lab coats walk briskly down a hospital corridor. A nurse uses a tablet while others carry patient charts and attend to a gurney. The setting conveys a busy, clinical environment focused on patient care.]
Use Cases

Why AI-native EHR platforms will treat speech as core infrastructure in 2026

As clinical workflows become automated and AI-driven, real-time speech is shifting from a transcription feature to the foundational intelligence layer inside modern EHR systems.

Vamsi Edara
Vamsi EdaraFounder and CEO, Edvak EHR
[alt: Logos of Speechmatics and Edvak are displayed side by side, interconnected by a stylized x symbol. The background features soft, wavy lines in light blue, creating a modern and tech-focused aesthetic.]
Company

One word changes everything: Speechmatics and Edvak EHR partner to make voice AI safe for clinical automation at scale

Turning real-time clinical speech into trusted, EHR-native automation.

Speechmatics
SpeechmaticsEditorial Team
[alt: Concentric circles radiate outward from a central orange icon with a white Speechmatics logo. The background is dark blue, enhancing the orange glow. A thin green line runs horizontally across the lower part of the image.]
Technical

Speed you can trust: The STT metrics that matter for voice agents

What “fast” actually means for voice agents — and why Pipecat’s TTFS + semantic accuracy is the clearest benchmark we’ve seen.

Archie McMullan
Archie McMullanSpeechmatics Graduate