The project was born from the desire to sing more than just one available song from a favorite band at karaoke.
The Web Speech API was chosen over paid services because it is a free, native browser feature for speech recognition and synthesis.
The API has significant limitations, including poor cross-browser support, reliance on vendor servers like Google and Apple, and privacy concerns.
Live demos showcase the API's core functions, including speech synthesis for reading selected text and speech recognition for voice navigation.
The game logic involves synchronizing lyrics with audio playback time and comparing the transcribed speech with the expected lyric text.
The completed karaoke game is demonstrated by speaking, singing, and intentionally using incorrect words to test the matching logic.
While the API is not yet production-ready for all use cases, it powers many interesting community projects, polyfills, and captioning tools.
When designing for voice, it's crucial to use concise content, consider conversational flow, and plan for error recovery and cognitive load.
Building "useless" or fun side projects is a powerful way to combat developer burnout, learn new technologies, and rediscover the joy of coding.
Sensory-Minds GmbH
Offenbach am Main, Germany
28:09
30:59Lee Boonstra
1:06:20Chris Heilmann, Daniel Cranney & Jan Deppisch
32:54Rowdy Rabouw
27:23Nico Martin
23:09Tobias Münch
14:44Chris Heilmann
1:13:35Chris Heilmann & Daniel Cranney & Rowdy Rabouw
Jobs that call for the skills explored in this talk.
Jobriver Hr Service
Berlin, Germany
Ai-native
Barcelona, Spain
