1 |
Some issues affecting the transcription of hungarian broadcast audio
|
|
|
|
In: Annual Conference of the International Speech Communication Association ; https://hal.archives-ouvertes.fr/hal-01843430 ; Annual Conference of the International Speech Communication Association , Aug 2013, Lyon, France (2013)
|
|
Abstract:
International audience ; This paper reports on a speech-to-text (STT) transcription system for Hungarian broadcast audio developed for the 2012 Quaero evaluations. For this evaluation, no manually transcribed audio data were provided for model training, however a small amount of development data were provided to assess system performance. As a consequence, the acoustic models were developed in an unsupervised manner, with the only supervision provided indirectly by the language model. The language models were trained on texts downloaded from various websites, also without any speech transcripts. This contrasts with other STT systems for Hungarian broadcast audio which use at least 10 to 50 hours of manually transcribed data for acoustic training, and typically include speech transcripts in the language models. Based on mixed results previously reported applying morph-based approaches to agglutinative languages such as Hungarian, word-based language models were used. The initial Word Error Rate (WER) of the system using context-independent seed models from other languages of 59.8% on the 3h development corpus was reduced to 25.0% after successive training iterations and system refinement. The same system obtained a WER of 23.3% on the independent Quaero 2012 evaluation corpus (a mix of broadcast news and broadcast conversation data). These results compare well with previously reported systems on similar data. Various issues affecting system performance are discussed, such as amount of training data, the acoustic features and choice of text sources for language model training.
|
|
Keyword:
[INFO.INFO-CL]Computer Science [cs]/Computation and Language [cs.CL]; [INFO]Computer Science [cs]; agglutinative languages; Bottleneck MLP features; broadcast news transcription; Hungarian language; Large vocabulary continuous speech recognition (LVCSR); unsupervised training
|
|
URL: https://hal.archives-ouvertes.fr/hal-01843430
|
|
BASE
|
|
Hide details
|
|
2 |
Acoustic unit discovery and pronunciation generation from a grapheme-based lexicon
|
|
|
|
In: IEEE Automatic Speech Recognition and Understanding Workshop ; https://hal.archives-ouvertes.fr/hal-01843433 ; IEEE Automatic Speech Recognition and Understanding Workshop, Dec 2013, Olomouc, Czech Republic (2013)
|
|
BASE
|
|
Show details
|
|
3 |
Discriminative training of a phoneme confusion model for a dynamic lexicon in ASR
|
|
|
|
In: Interspeech 2013 ; Annual Conference of the International Speech Communication Association ; https://hal.archives-ouvertes.fr/hal-01843427 ; Annual Conference of the International Speech Communication Association, Jan 2013, Lyon, France (2013)
|
|
BASE
|
|
Show details
|
|
4 |
Recent evolution of non-standard consonantal variants in French broadcast news
|
|
|
|
In: Interspeech ; https://halshs.archives-ouvertes.fr/halshs-00856290 ; Interspeech, Aug 2013, Lyon, France. pp.412-416 (2013)
|
|
BASE
|
|
Show details
|
|
5 |
Recent Evolution of Non Standard Consonantal Variants in French Broadcast News
|
|
|
|
In: Annual Conference of the International Speech Communication Association ; https://hal.archives-ouvertes.fr/hal-01843431 ; Annual Conference of the International Speech Communication Association , International Speech Communication Association, F. Bimbot, C. Cerisara, C. Fougeron, G. Gravier, L. Lamel, F. Pellegrino, P. Perrier, Jan 2013, Lyon, France (2013)
|
|
BASE
|
|
Show details
|
|
6 |
Unsupervised Acoustic Model Training with Limited Linguistic Resources
|
|
|
|
In: IEEE Automatic Speech Recognition and Understanding Workshop ; https://hal.archives-ouvertes.fr/hal-01843476 ; IEEE Automatic Speech Recognition and Understanding Workshop, Jan 2013, Olomouc, Czech Republic (2013)
|
|
BASE
|
|
Show details
|
|
7 |
What we can learn from ASR errors about low-resourced languages: a case- study of Luxembourgish and Austrian
|
|
|
|
In: Errors by Humans and Machines in Multimedia, Multimodal, Multilingual Data Processing ; https://hal.archives-ouvertes.fr/hal-01843440 ; Errors by Humans and Machines in Multimedia, Multimodal, Multilingual Data Processing, Jan 2013, Ermenonville, France (2013)
|
|
BASE
|
|
Show details
|
|
8 |
Embosi: automatic alignment with segments and words and phonological mining
|
|
|
|
In: International Conference on Bantu Languages ; https://hal.archives-ouvertes.fr/hal-01843438 ; International Conference on Bantu Languages, Jan 2013, Paris, France (2013)
|
|
BASE
|
|
Show details
|
|
9 |
What we can learn from asr errors about low-resourced languages: a case-study of luxembourgish and austrian
|
|
|
|
In: Errors by Humans and Machines in Multimedia, Multimodal, Multilingual Data Processing (ERRARE 2013) ; https://halshs.archives-ouvertes.fr/halshs-01424902 ; Errors by Humans and Machines in Multimedia, Multimodal, Multilingual Data Processing (ERRARE 2013), Nov 2013, Ermenonville, France (2013)
|
|
BASE
|
|
Show details
|
|
10 |
Embosi : automatic alignment with segments and words and phonological mining
|
|
|
|
In: International Conference on Bantu Languages (BANTU 2013) ; https://halshs.archives-ouvertes.fr/halshs-01424894 ; International Conference on Bantu Languages (BANTU 2013), Jun 2013, Paris France (2013)
|
|
BASE
|
|
Show details
|
|
11 |
Human annotation of asr error regions: Is ”gravity” a sharable concept for human annotators?
|
|
|
|
In: Errors by Humans and Machines in Multimedia, Multimodal, Multilingual Data Processing (ERRARE 2013) ; https://halshs.archives-ouvertes.fr/halshs-01424915 ; Errors by Humans and Machines in Multimedia, Multimodal, Multilingual Data Processing (ERRARE 2013), Nov 2013, Ermenonville, France (2013)
|
|
BASE
|
|
Show details
|
|
12 |
Systèmes de transcription comme instruments
|
|
|
|
In: Méthodes et outils pour l'analyse phonétique des grands corpus oraux ; https://hal.archives-ouvertes.fr/hal-01135113 ; Nguyen Noël; Adda-Decker Martine. Méthodes et outils pour l'analyse phonétique des grands corpus oraux, Hermes Science Publications, pp.159-202, 2013, Cognition et Traitement de l'Information, 978-2746245303 (2013)
|
|
BASE
|
|
Show details
|
|
13 |
Proceedings of the 14th Annual Conference of the International Speech Communication Association (Interspeech), 25-29 August 2013, Lyon (France)
|
|
|
|
In: https://hal.archives-ouvertes.fr/hal-00931864 ; France. International Speech Communication Association (ISCA), over 3500 p., 2013 (2013)
|
|
BASE
|
|
Show details
|
|
|
|