Nowadays, speech technology permeates our everyday life. From smart-home devices to in-car navigation systems to chatting with friends on the other side of the planet, countless speech applications inform our worlds and enable us to do things that were unthinkable only a few years ago and that we now take for granted.
However, in thinking about how this technology has changed the lives of so many, it’s worth recalling the famous quote from science fiction writer William Gibson: “The future is already here — it’s just not very evenly distributed.”
While a native English speaker can enjoy a wide set of technologies with state-of-the-art performances, the same might not be true for a Balinese or a Gulf Arabic speaker. We want to help change that.
The Spoken Language Technology Workshop (SLT) is the biannual flagship event of the IEEE Speech and Language Processing Technical Committee. As such, SLT brings together researchers from both academia and industry to present their work and learn about state-of-the-art technologies in fields such as automatic speech recognition, text-to-speech, spoken dialogue systems, and speech-to-speech translation.
Next year’s event will be significant, as it will be the first time a speech technology conference takes place in the Middle East and in an Arabic-speaking nation. In celebration of these milestones, the theme of this year’s event is “Languages of the world”.
Moreover, for the first time in the history of SLT, we are organizing SLT-Code, a two-day speech technology hackathon. Following the main workshop’s theme, the hackathon will focus on language diversity. Our goal: develop a diverse community of speech-technology practitioners and researchers, specifically targeting the inclusion of researchers from underrepresented regions and low-resource languages.
This year’s areas of interest for the hackathon include (but are not limited to)
- Low-resource languages, e.g., African, Indian, or Asian languages
- Language identification
- Automatic speech recognition
- Machine translation
- Text to speech
- Speech-to-speech translation
While participants will be free to use any technology of their preference, the organizers will support the following frameworks:
- Lex/Alexa
- Huggingface
- Speechbrain
- ESPNet
- K2/Kaldi
We are looking forward to welcoming participants from different backgrounds and are eager to read about your speech technology ideas. To participate, please register by September 30, 2022.
Important dates
- September 30, 2022: Call for applications and submission of idea proposals
- October 12, 2022: Idea selection and invitation to participate
- October 12 – November 1, 2022: Team-forming phase, mentor assignment
- January 7 – 8, 2023: SLT-Code
Organizers
Thomas Schaaf (3M | M*Modal)
Gianni Di Caro (Carnegie Mellon University Qatar)
Shinji Watanabe (ESPNET | Carnegie Mellon University)
Paola Garcia (KALDI/K2 | Johns Hopkins University)
Mirco Ravanelli (SpeechBrain | Université de Montréal)
Alessandra Cervone (Amazon Alexa AI)
Mus’ab Husaini (QCRI)
Harshita Diddee (Microsoft)
