YouTube—YouTube launched automatic captions using speech recognition, expanding video accessibility for deaf and hard-of-hearing users
In November 2009, YouTube introduced automatic captions using speech recognition technology, initially for English-language content. The feature was led by Ken Harrenstien, a deaf Google engineer. While auto-captions had significant accuracy limitations (60-70% accuracy initially, improving over time), the feature represented a major step toward making the platform's massive video library accessible to deaf and hard-of-hearing users. YouTube subsequently expanded auto-captions to dozens of languages.
Scoring Impact
| Topic | Direction | Relevance | Contribution |
|---|---|---|---|
| Accessibility | +toward | primary | +1.00 |
| Overall incident score = | +0.885 | ||
Score = avg(topic contributions) × significance (high ×1.5) × confidence (0.59)
Evidence (1 signal)
YouTube launched automatic speech recognition captions for accessibility
In November 2009, YouTube introduced automatic captions using speech recognition technology, initially for English-language educational content. The project was led by Ken Harrenstien, a deaf Google engineer who said he 'would be ecstatic to see even the most inaccurate captions.' While initial accuracy was around 60-70%, the feature represented a significant accessibility advance for the platform's massive video library.