Indian languages offer a multitude of challenges that are not observed as widely in languages elsewhere, e.g. code-switching ("Hinglish"). However, research on ASR for Indian languages remains relatively scarce compared to English, European languages, and East-Asian languages.

We will invite participants of this Special Session to build their own ASR systems for Indian languages, in the broad sense of the term. You may choose which Indian languages to build systems for, and what data sets to use.

In addition to the challenge around building ASR systems, we will also allocate time in the schedule for a show-and-tell: some participants may prefer using an existing ASR system and then building a voice-enabled app.

We believe that this Special Session would provide an important boost to academic research on building ASR for Indian languages.