%Aigaion2 BibTeX export from Idiap Publications
%Thursday 21 November 2024 12:40:04 PM

@INPROCEEDINGS{Dines_INTERSPEECH-2_2009,
         author = {Dines, John and Saheer, Lakshmi and Liang, Hui},
       keywords = {decision trees, speech recognition, speech synthesis, unified models},
       projects = {EMIME},
          month = {9},
          title = {Speech recognition with speech synthesis models by marginalising over decision tree leaves},
      booktitle = {Proceedings of Interspeech},
           year = {2009},
       location = {Brighton, U.K.},
       crossref = {Dines_Idiap-RR-17-2009},
       abstract = {There has been increasing interest in the use of unsupervised adaptation for the personalisation of text-to-speech (TTS) voices, particularly in the context of speech-to-speech translation. This requires that we are able to generate adaptation transforms from the output of an automatic speech recognition (ASR) system. An approach that utilises unified ASR and TTS models would seem to offer an ideal mechanism for the application of unsupervised adaptation to TTS since transforms could be shared between ASR and TTS. Such unified models should use a common set of parameters. A major barrier to such parameter sharing is the use of differing contexts in ASR and TTS. In this paper we propose a simple approach that generates ASR models from a trained set of TTS models by marginalising over the TTS contexts that are not used by ASR. We present preliminary results of our proposed method on a large vocabulary speech recognition task and provide insights into future directions of this work.},
            pdf = {https://publications.idiap.ch/attachments/papers/2009/Dines_INTERSPEECH-2_2009.pdf}
}



crossreferenced publications: 
@TECHREPORT{Dines_Idiap-RR-17-2009,
         author = {Dines, John and Saheer, Lakshmi and Liang, Hui},
       projects = {EMIME},
          month = {7},
          title = {Speech recognition with speech synthesis models by marginalising over decision tree leaves},
           type = {Idiap-RR},
         number = {Idiap-RR-17-2009},
           year = {2009},
    institution = {Idiap},
       abstract = {There has been increasing interest in the use of unsupervised 
adaptation for the personalisation of text-to-speech (TTS) 
voices, particularly in the context of speech-to-speech translation.
This requires that we are able to generate adaptation 
transforms from the output of an automatic speech recognition 
(ASR) system. An approach that utilises unified ASR and TTS 
models would seem to offer an ideal mechanism for the application
of unsupervised adaptation to TTS since transforms could
 be shared between ASR and TTS. Such unified models should 
use a common set of parameters. A major barrier to such parameter 
sharing is the use of differing contexts in ASR and TTS. In
 this paper we propose a simple approach that generates ASR
 models from a trained set of TTS models by marginalising over
the TTS contexts that are not used by ASR. We present preliminary
 results of our proposed method on a large vocabulary
 speech recognition task and provide insights into future directions 
of this work.},
            pdf = {https://publications.idiap.ch/attachments/reports/2009/Dines_Idiap-RR-17-2009.pdf}
}