Behind The Scenes: A Look Back At A Month Of Real-World AI API Latency At Scale
Last week we examined a 24-hour period of real-world AI API latency and error rates as an illustration of the…
Behind The Scenes: GCP's Network Intelligence Performance Dashboard
The sheer massiveness of GCP's core service offerings is such that there are a wealth of hidden gems buried within…
Behind The Scenes: GCP Network Intelligence Topology Mapping Of Our OCR Cluster At Startup
GCP's Network Intelligence service offers an incredibly powerful Network Topology visualization that shows all of the various GCP services being…
Behind The Scenes: Using Our Bigtable + BigQuery + GCS Digital Twin To Queue Missing Broadcasts
Our massive new collaboration with the Internet Archive to OCR its complete quarter-century Television News Archive spans ten million broadcasts…
At-Scale OCR Of Television News Experiments: Ornamental Vs Chyron Text
One of the great tradeoffs in using image montaging to achieve a 100-200x performance increase and cost reduction for at-scale…
At-Scale OCR Of Television News Experiments: Vertical + Horizonal Text – Results From A Taiwanese Broadcast
Unlike most of the world, television news broadcasts in Taiwan make regular use of vertical onscreen text that appears alongside…
At-Scale OCR Of Television News Experiments: Results From A Sample Circa-2018 South Sudan Broadcast
Below is a transcribed OCR excerpt from a circa-2018 South Sudanese broadcast. As with our previous examples, the difficulties of…
Behind The Scenes: Managing The Unpredictability Of Cloud AI API Latency At Archive Scale
While all APIs can exhibit variable response times and error rates, AI APIs demonstrate uniquely complex response behaviors. The scarcity…
At-Scale OCR Of Television News Experiments: Results From A Sample Circa-2018 Congolese Broadcast – Part 2
In contrast to yesterday's Congolese television news example, in which Cloud Vision API was unable to transcribe the onscreen chyron…
At-Scale OCR Of Television News Experiments: Results From A Sample Circa-2018 Congolese Broadcast
While our Cloud Vision montaging workflow has yielded highly robust results across the majority of broadcasts we've tested it on,…
At-Scale OCR Of Television News Experiments: Results From A Sample Circa-2018 Nigerian Broadcast
Below is an OCR excerpt from a circa-2018 Nigerian news broadcast, with a fast-paced white textual "crawl" over a red…
At-Scale OCR Of Television News Experiments: Results From A Sample Circa-2012 Venezuelan Broadcast
How does our OCR workflow perform on a circa-2012 Venezuelan broadcast? Of interest, it even captures the "Reuters" byline of…
At-Scale OCR Of Television News Experiments: Results From A Sample Circa-2012 Vietnamese Broadcast
Continuing our OCR series, below is a circa-2012 Vietnamese broadcast. In keeping with the highly multilingual world of global broadcast…
At-Scale OCR Of Television News Experiments: Results From A Sample Circa-2018 Amharic Broadcast – Part 2: Onscreen Documents
Continuing yesterday's examination of OCRing an Amharic-language broadcast, towards the end of that broadcast is a fascinating example of a…
At-Scale OCR Of Television News Experiments: Results From A Sample Circa-2018 Amharic Broadcast – Part 1
Continuing our OCR experiments applying GCP's Cloud Vision API to global television news broadcasts using montaging, below is an excerpt…
At-Scale OCR Of Television News Experiments: What Have We Learned So Far?
In collaboration with the Internet Archive's TV News Archive, we are working to OCR the Archive's entire 7-million-hour quarter-century archive…
Behind The Scenes: The Perils Of AI-Powered Autonomous Agents In The Real World
We continue to explore the landscape of AI-powered autonomous agents. Despite their immense hype and ubiquitous mediagenic demos on social…
GCP Tips & Tricks: Using The Cloud Monitoring API To Track AI API Usage In Realtime
Yesterday we discussed how we massively optimized our archive-scale OCR throughput by splitting montaging and OCR workloads. When working at the…
Behind The Scenes: Splitting Workloads: OCR Montage Generation Vs API Calls
As we continue to scale up our work OCR'ing a quarter century of global television news broadcasts, one of the…
At-Scale OCR Of Television News Experiments: Results From A Tunisian Broadcast
Below is an example of our Cloud Vision montage pipeline's transcription of a Tunisian broadcast from 2014, showing how it…
At-Scale OCR Of Television News Experiments: Results From A Turkish TRT 1 Broadcast
Continuing our Cloud Vision television news OCR montaging experiments, below is a circa-2012 example from Turkish broadcaster TRT 1, showing…
At-Scale OCR Of Television News Experiments: Results From A Thai Broadcast
Continuing our Cloud Vision television news OCR montaging experiments, below is a circa-2012 example of a Thai television news broadcast, showing…
At-Scale OCR Of Television News Experiments: Remarkable Results From A Circa-2012 320×240 Pixel Syrian Broadcast
Continuing our Cloud Vision television news OCR montaging experiments, below is a truly remarkable circa-2012 example from Syrian television. The…
At-Scale OCR Of Television News Experiments: Results From A Sample Quarter-Century-Old Chinese Language Broadcast
Thus far, we've demonstrated the results of applying GCP Cloud Vision to OCR a grid montage of 1fps frames from…