Case Study - Media Track

How Media Track unlocked a 20% leap in accuracy

Learn how Media Track, a global leader in media intelligence, improved their transcription accuracy with Speechmatics’ real-time technology.

The Organization

Media Track transforms audio-visual and print content into structured, searchable data for the media intelligence industry.

Operating across six continents and in 20+ languages, they monitor over 2 million print pages and over 3,000 channels globally of broadcast content every month.

With innovation at their core, Media Track delivers scalable, accurate solutions, continuously adopting cutting-edge technology as a member of FIBEP and AMEC.

The Challenge

Before partnering with Speechmatics, Media Track faced a core challenge: balancing high-quality and customization with the need for scalability and speed.

They needed to centralize their speech-to-text (STT) engines while maintaining accuracy and covering a wide variety of languages for their global clients.

The challenge was particularly complex because multiple STT engines were required to handle different language regions.

Media Track sought a solution that could centralize their operations, maintain high accuracy across multiple languages and scale effectively while reducing complexity

The Solution

In 2023, Media Track began testing Speechmatics and quickly realized the potential of integrating Speechmatics’ speech-to-text technology into their MT product.

After rigorous testing across languages such as Danish, Mandarin, and German, Media Track saw an impressive 20% improvement in overall speech-to-text accuracy.

Key reasons for choosing Speechmatics included:

  • Wide language coverage in real-time Speechmatics supported the wide variety of languages required for Media Track’s global operations.

  • Accuracy and speed Speechmatics demonstrated faster processing and higher accuracy in transcription compared to other engines.

  • Scalability Centralizing with Speechmatics simplified workflows and made scaling easier, ensuring a more efficient media monitoring process.

The Impact

By integrating Speechmatics, Media Track achieved a 20% improvement in speech-to-text accuracy.

Media Track saw a 20% improvement in overall accuracy of the speech-to-text (based on Q4 2023 - Q1 2024) tests.

Language

Speechmatics WER

Google WER

Chinese

20.8%

40.5%

Danish

19.7%

41.7%

Dutch

26.9%

28.8%

English

5.6%

24.5%

Finnish

18.2%

39.8%

French

10.4%

20.0%

German

23.1%

48.4%

Indonesian

42.7%

62.5%

Italian

11.5%

21.7%

Malay

19.9%

40.0%

Norwegian

21.1%

40.0%

Russian

16.4%

47.4%

Swedish

33.1%

63.9%

This enhancement has allowed them to:

  • Improve transcription performance across more than 20 languages, ensuring higher accuracy and reliability for their global clients.

  • Increase operational efficiency by streamlining multiple workflows, leading to faster and more scalable solutions.

  • Offer enhanced services to clients, such as optional music detection, further expanding Media Track’s media monitoring capabilities.

What's next?

Looking forward, Media Track plans to continue leveraging Speechmatics’ solutions, including exploring new functionalities that align with the latest trends in AI and speech recognition.

With features like bilingual languages and improved audio events (music detection and identification), Media Track anticipates further enhancing its own AI services (speech, music, segmentation, entities, logo, and OCR) and media monitoring offerings.

The partnership between Speechmatics and Media Track will continue to evolve, with both companies collaborating on new features and innovations to stay ahead in the media intelligence space.

"Speechmatics delivers continuous accuracy improvements across multiple languages, allowing us to provide timely transcription with incredibly precision. Their technology has become essential to our global media monitoring operations."

Raphael QuisumbingHead of Product, Media Track
Media Track quote
Captioning

Limecraft

How Limecraft utilizes AI for generating transcripts and automating workflows.

CCaaS

Liquid Voice

How Liquid Voice launched a ground-breaking solution, integrating ASR technology from Speechmatics.

Captioning

aiconix

How aiconix generate live subtitles within seconds and improve accessibility.

Live. Instant. Real-time.

Precise, low-latency transcription and translation, all delivered before your media even ends. Why wait, when you can deliver incredible features now?