Initially, we planned on creating an Android application to act as a microphone for recording user audio to analyze. Speech recognition was a key part in this project, however it was also one of its most challenging.Based on the information calculated, the Pebble smartwatch will display the appropriate words on its screen (a few words ahead of where the user currently is). This is accomplished using fuzzy string analysis, specifically mathematical models and methods e.g. These files are analyzed asynchronously by Google's Speech API to determine the position of the user within the speech's script. PyAudio, a module, is used to record RAW audio files of 3 seconds constantly. In order to determine the progress of the speech, it analyzes the voice of the user, nearly in real-time. How we built itĪt its heart, Pebble Prompter is an automatic scrolling teleprompter. Also, lecture supervisors or teachers can send comments through our Python interface such as "5 minutes left!" accompanied with haptic feedback to the user's watch. At any time, if the user needs assistance to remember the next part of their speech, they can look at the screen of their Pebble watch. Throughout the speech, the program will constantly monitor the user's voice in order to determine how far the user has progressed in their speech. Initially, the user will enter the script of a speech which will be used for dynamically finding the position of the user in the speech. Essentially, our computer program analyzes speech samples from the user periodically using an in-built or external microphone. We realized that this system could be replaced by an automated system using speech recognition techniques in order to reduce the amount of human effort/time spent manually controlling teleprompters. These prompters are controlled by humans, allowing for variations in the speed of the teleprompter to enable the news anchor to speak naturally during the newscast. Pebble prompter aims to simulate prompters offered to news anchors in major news corporations e.g. To resolve this problem, we have created Pebble Prompter. A potential solution is to record the speech and play it through an earpiece as an audio file, however this method will not compensate for pauses or diversions during the speech, which are natural. Generally, this occurs when the speech is quite large in length. Personally, we have seen many fellow classmates stumble and forget parts of their desired scripts/dialogues when presenting in front of large audiences. Often, during school time we are asked to give presentations or lectures in activities/classes. Composed of a diverse team, consisting of both high-school and university students, we strived to create an idea that would help us in our academic endeavours.