Sherlock Holmes uses IBM Watson on Bluemix instead of Dr. Watson

Share this post:

Picture this: You walk up to a chalk outlined body. There are several items around the body – a locket with a photo, a cigarette case, and a lock of hair. A 19th century rotary phone starts ringing next to you. You pick up the phone and you hear an automated voice say “Thank goodness someone picked up the phone! We’ve been trying to get through for hours! Crime scenes are falling apart all over London, and Scotland Yard can’t handle all the calls. We desperately need your help, Detective. What’s your name and Precinct?”

You speak into the phone and the automated voice asks you what clues you have next to the body. Then you speak back and forth with this mysterious voice about the clues to try to understand who was the killer and piece together this mysterious crime. You are the modern day Sherlock Holmes with IBM Watson as your assistant – not Dr. Watson.

This is a social experiment constructed through a series of hackathons with Columbia University and the Made in NY Media Center. It combines the Internet of Things (IoT), Watson APIs, and IBM Bluemix to create a compelling social experience for strangers to construct their own Sherlock Holmes inspired story.

sherlock holmes collage

Hackathons bring together creatives, designers, developers, and Sherlock Holmes enthusiasts to stitch together this experience.

How Sherlock Holmes uses IBM Watson on Bluemix

Behind the scenes, a Raspberry PI is hooked up to a rotary phone which connects to a Bluemix application on the cloud. The Raspberry PI rings and then prompts the user to walk to a certain area to investigate clues around a chalk outlined body to solve a crime scene. Using beacons that are placed in a public space and based on the game time, the Raspberry PI calls the Watson services to understand what to say and when to speak. The application uses Watson’s speech-to-text service to translate the user’s voice into text to feed into the conversation service that creatives can use to create a compelling narrative. Then, responses from the conversation service feed into the text-to-speech service that plays out of the speaker to the rotary phone’s earpiece.

The current project is nearing completion this November 2016, with hopes of showcases out to the general public and at IBM events in 2017. There’s been a numerous amount of support from IBMers and the general hackathon and startup communities. Using Watson and IoT, we are now able to drive an entire brand new social way for people to interact and construct a virtual crime scene using Sherlock Holmes. We hope this project continues on and extends to different adaptions that can be used for underprivileged children and global Sherlock Holmes hackathons around the world.


Johnson Liu is a software developer at IBM in the NYC Bluemix Garage. Previously, he was a software engineer in the Enterprise Content Management group of IBM. Johnson is an author of 3 IBM RedBooks and over 10 developerWorks articles.

Cloud Garage Developer

More Community stories
April 29, 2019

Transforming Customer Experiences with AI Services (Part 1)

This is an experience from a recent customer engagement on transcribing customer conversations using IBM Watson AI services.

Continue reading

April 26, 2019

Updated Tutorial: Database-Driven Chatbot

The tutorial on how to build a database-driven chatbot has been updated. It's now simpler to deploy and and offers more options—Slack, Facebook Messenger, Wordpress, and more.

Continue reading

April 23, 2019

Announcing the Deprecation of the Watson Machine Learning JSON Token Authentication Service

We’d like to inform you about the deprecation of the Watson Machine Learning JSON Token Authentication service. This method of authentication will be retired on May 30, 2019.

Continue reading