Domain adaptation for speech and language processing

The Idiap Research Institute in partnership with Swisscom invites applications for a post-doctoral (or similarly qualified) position in automatic speech recognition (ASR) and natural language processing (NLP). The position is funded by Swisscom, with a view to a long term collaboration.

The successful candidate will work with Dr. Philip Garner and/or Dr. James Henderson, and with the wider Speech and Audio Processing and Natural Language Processing groups, engaged in world-class research in speech and language processing. Exceptionally qualified candidates can also be considered for a longer-term Research Associate position.

Description
Idiap will address new emerging challenges related to fast customisation of ASR and NLU technologies. This is particulary important where interaction with customers can be for short times over channels of various quality and over different modes.

The research is likely to build on current state of the art end-to-end techniques enabling domain adaptation of both the ASR and NLP models. We aim to focus initially on the lexicon, aiming to propagate advances in open vocabulary NLP (such as machine translation) into ASR.

The successful candidate will be based at Idiap in Martigny, but will work in close collaboration with Swisscom's R&D teams based in Lausanne and elsewhere in Switzerland. This means spending one or more days per week at the Swisscom Dital lab on the EPFL campus. The projects are a combination of applied science and academic research; as well as generating research papers, there will be the opportunity to propagate the results of the research into real applications at Swisscom.

Profile
Candidates should have either or both of:
1. A strong background in engineering, mathematics or a related discipline, along with the associated familiarity with modern distributed programming environments and languages such as C++, Python and Perl.
2. An exceptional academic record and a clear aptitude for creative (and independent) research in a related discipline.
In either case, familiarity with speech processing tools (e.g., Kaldi), natural language processing tools, or deep learning toolkits (e.g., pyTorch, tensorflow) will be a distinct advantage. Although a PhD is normally a prerequisite for a post-doctoral position, candidates without a PhD may be considered in exceptional cases.

Timescale
All positions are offered on a one-year basis with the possibility of renewal based on funding and performance. The starting salary will be 80,000 CHF/year. Starting date is as soon as possible, modulo permit regulations...

Interested? Please submit your candidature through the Idiap online recruitment system: Domain adaptation for speech and language processing