forum | IKS - Details
forum | IKS: Vortrag von Prof. Dr.‐Ing. Tim Fingscheidt
Institut für Nachrichtentechnik, TU Braunschweig
Turbo Automatic Speech Recognition
Mittwoch, 28. Juni 2017
The event series will be continued with a lecture by Prof. Dr.‐Ing. Tim Fingscheidt from TU Braunschweig entitled:
„Turbo Automatic Speech Recognition”.
Performance of automatic speech recognition (ASR) systems can significantly be improved by integrating further sources of information such as additional modalities, acoustic channels, or acoustic models. Given the arising problem of information fusion, striking parallels to problems in digital communications are exhibited, where the discovery of the turbo codes by Berrou et al. was a groundbreaking innovation. In this presentation, we show ways how to successfully apply the turbo principle to the domain of ASR and thereby provide solutions to the information fusion problem. We will present the turbo decoding forward‐backward algorithm (FBA), giving detailed insights into turbo ASR. Then, we present simulation results for a multimodal example of information fusion. Finally, we prove the suitability of both our turbo ASR approaches also for a single‐channel multimodel recognition. On a small vocabulary task (challenging, since spelling is included), our proposed turbo ASR approach outperforms even the best reference system on average over all SNR conditions and investigated noise types by a relative word error rate reduction of 22.4% (audio‐visual task) and 18.2% (audio‐only task), respectively.
After the lecture, the guests have the opportunity to join an open discussion during a snack.