Our goal was to implement a few different games using the speech to text API provided by rev.ai and to support it with a UI. The games we implemented were Caveman, Extra, and UpDown. All games make use of the speech to text API. Caveman was our first game and it involves the users inputting a number, such as 500, which determined how many words are used. Player 1 then describes a random item to player 2 using only what they believe are the 500 most frequently used words in the English language. The API recognizes words outside of this set and buzzes when an invalid word is used. When player 2 guesses the word, the turn ends and time is recorded.
Extra is the opposite: it requires the usage of infrequently used words to describe the random item. Higher point values are assigned to words that are seldom used, so there is a benefit to using less common words.
Our final game was Up Down. A single user is given a randomly chosen word from the 60,000 most frequently used. They are then asked to say a word that is higher or lower than the one given. The choice between higher or lower is random. The word spoken by the user is then used for the next round and it repeats. A high score is recorded for the future.
The biggest problem we faced was the integration of all the separate components; this proved to be the most daunting task. Early design planning would have mitigated this.
Built With
- pyqt5
- python
- rev.ai
Log in or sign up for Devpost to join the conversation.