{"id":141,"date":"2019-04-27T22:13:23","date_gmt":"2019-04-27T22:13:23","guid":{"rendered":"http:\/\/course.ece.cmu.edu\/~ece500\/projects\/s19-teamd2\/?p=141"},"modified":"2019-05-06T00:17:23","modified_gmt":"2019-05-06T00:17:23","slug":"status-report-10-apr-27","status":"publish","type":"post","link":"http:\/\/course.ece.cmu.edu\/~ece500\/projects\/s19-teamd2\/2019\/04\/27\/status-report-10-apr-27\/","title":{"rendered":"Status Report #10 (Apr 27)"},"content":{"rendered":"

Haohan Shi<\/h2>\n

This week I mainly worked on improvement and assembly of our case. The initial design of our outer case lacks mounting\u00a0mechanics and relies only on mechanical\u00a0hooks, and it turns out to be fairly unstable since the 3D printing isn’t accurate enough, and the laser-cut material such as wood or acrylic sheets are too heavy and thick. In addition, laser-cutting cannot make curvature\u00a0surfaces easily so that the initial design doesn’t look very nice when we assembled.<\/p>\n

For our second edition, we added the mounting holes which matched the screws holes on the robot platform and used glue guns to seal all the connection parts.<\/p>\n

Then I conducted several tests with my teammates\u00a0to find out what parameters work best for our hot word\u00a0detection.<\/p>\n

 <\/p>\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
Sensitivity<\/span><\/td>\nGain<\/span><\/td>\nDistance (from mic)<\/span><\/td>\nNoise<\/span><\/td>\nFalse Alarm time<\/b><\/td>\nHotword Success <\/b><\/td>\nSpeech-Recognition Success<\/b><\/td>\n<\/tr>\n
0.4<\/span><\/td>\n1<\/span><\/td>\n5<\/span><\/td>\nNone<\/span><\/td>\n0<\/span><\/td>\n0%<\/span><\/td>\n–<\/span><\/td>\n<\/tr>\n
0.4<\/b><\/td>\n1<\/b><\/td>\n10<\/b><\/td>\nNone<\/b><\/td>\n1<\/b><\/td>\n66.7%<\/b><\/td>\n100%<\/b><\/td>\n<\/tr>\n
0.4<\/b><\/td>\n1<\/b><\/td>\n15<\/b><\/td>\nNone<\/b><\/td>\n0<\/b><\/td>\n80%<\/b><\/td>\n62.5%<\/b><\/td>\n<\/tr>\n
0.4<\/span><\/td>\n1<\/span><\/td>\n25<\/span><\/td>\nNone<\/span><\/td>\n0<\/span><\/td>\n0%<\/span><\/td>\n–<\/span><\/td>\n<\/tr>\n
0.5<\/span><\/td>\n1<\/span><\/td>\n5<\/span><\/td>\nNone<\/span><\/td>\n4<\/span><\/td>\n50%<\/span><\/td>\n50%<\/span><\/td>\n<\/tr>\n
0.5<\/span><\/td>\n1<\/span><\/td>\n15<\/span><\/td>\nNone<\/span><\/td>\n3<\/span><\/td>\n10%<\/span><\/td>\n100%<\/span><\/td>\n<\/tr>\n
0.5<\/span><\/td>\n1<\/span><\/td>\n25<\/span><\/td>\nNone<\/span><\/td>\n0<\/span><\/td>\n10%<\/span><\/td>\n100%<\/span><\/td>\n<\/tr>\n
0.4<\/span><\/td>\n3<\/span><\/td>\n5, 15<\/span><\/td>\nNone<\/span><\/td>\n0<\/span><\/td>\n0%<\/span><\/td>\n–<\/span><\/td>\n<\/tr>\n
0.4<\/b><\/td>\n3<\/b><\/td>\n25<\/b><\/td>\nNone<\/b><\/td>\n0<\/b><\/td>\n71.4%<\/b><\/td>\n70%<\/b><\/td>\n<\/tr>\n
0.4<\/b><\/td>\n3<\/b><\/td>\n35<\/b><\/td>\nNone<\/b><\/td>\n0<\/b><\/td>\n90%<\/b><\/td>\n66.7%<\/b><\/td>\n<\/tr>\n
0.4<\/span><\/td>\n3<\/span><\/td>\n45<\/span><\/td>\nNone<\/span><\/td>\n0<\/span><\/td>\n0%<\/span><\/td>\n–<\/span><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

<\/h2>\n\n\n\n\n\n\n\n\n\n\n
Sensitivity<\/span><\/td>\nGain<\/span><\/td>\nDistance (from mic)<\/span><\/td>\nNoise<\/span><\/td>\nFalse Alarm time<\/b><\/td>\nHotword Success <\/b><\/td>\nSpeech-Recognition Success<\/b><\/td>\n<\/tr>\n
0.4<\/span><\/td>\n5<\/span><\/td>\n5, 15<\/span><\/td>\nNone<\/span><\/td>\n1<\/span><\/td>\n0%<\/span><\/td>\n–<\/span><\/td>\n<\/tr>\n
0.4<\/span><\/td>\n5<\/span><\/td>\n25<\/span><\/td>\nNone<\/span><\/td>\n1<\/span><\/td>\n29.4%<\/span><\/td>\n71.4%<\/span><\/td>\n<\/tr>\n
0.4<\/span><\/td>\n3<\/span><\/td>\n5, 15<\/span><\/td>\nNoisy<\/span><\/td>\n0<\/span><\/td>\n0%<\/span><\/td>\n–<\/span><\/td>\n<\/tr>\n
0.4<\/b><\/td>\n3<\/b><\/td>\n25<\/b><\/td>\nNoisy<\/b><\/td>\n0<\/b><\/td>\n66.7%<\/b><\/td>\n58.4%<\/b><\/td>\n<\/tr>\n
0.4<\/span><\/td>\n3<\/span><\/td>\n35<\/span><\/td>\nNoisy<\/span><\/td>\n0<\/span><\/td>\n38.5%<\/span><\/td>\n70%<\/span><\/td>\n<\/tr>\n
0.4<\/span><\/td>\n3<\/span><\/td>\n5, 15, 35<\/span><\/td>\nVery Noisy<\/span><\/td>\n0<\/span><\/td>\n0%<\/span><\/td>\n0%<\/span><\/td>\n<\/tr>\n
0.4<\/span><\/td>\n3<\/span><\/td>\n25<\/span><\/td>\nVery Noisy<\/span><\/td>\n0<\/span><\/td>\n9.1%<\/span><\/td>\n100%<\/span><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

The result provides a lot of insights that we didn’t think of initially. It turns out 0.4 and 3 are optimal parameters currently because the ambient noise may be very loud and we are ~25 inches away from the robot during the demo. But more data needs to be collected during a noisy and very noisy environment.<\/p>\n

Olivia Xu<\/h2>\n

Designed new case for the robot with all the information from test runs from last week. Need a good way to mount things and need larger space. So I made a bottom with tabs that have hole for mounting with screw onto the robot, this also allows a much larger design above. Surprisingly weight isn’t much of a problem when the speed is turned up.<\/p>\n

Notable constraints:<\/p>\n