This paper presents a speech signal transformation which slows down speech signals selectively and enhances some important acoustic cues. This transformation can be used for hearing aids as well as second language acquisition by facilitating oral comprehension. Selective slowing down relies on the use the TD-PSOLA synthesis method. The strategy used to control slowing down exploits a spectral variation function which locates rapid spectral changes. The enhancement simply consists of amplifying stop bursts and unvoiced fricatives. These acoustic cues are detected automatically through the examination of energy criteria. Unvoiced transitions are also amplified. This approach was evaluated in the context of second language acquisition, more precisely by evaluating improvements in oral comprehension through the transformation, the detection and a perceptive experiment. Evaluation carried out over 50 french sentences. Each listener (over 50 non-native adults) had to complete missing words in a written transcription of the sentence they listened to. These listeners were learning french as second (or third) language for two years. Transformations were performed over the whole of each sentence (not only missing words). The detection of relevant events gives good results and experiments show that the oral comprehension is improved.

EXPLOR_STEP=$WICRI_ROOT/Wicri/Lorraine/explor/InforLorV4/Data/Main/Exploration

HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 009577 | SxmlIndent | more

HfdSelect -h $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd -nk 009577 | SxmlIndent | more

{{Explor lien
   |wiki=    Wicri/Lorraine
   |area=    InforLorV4
   |flux=    Main
   |étape=   Exploration
   |type=    RBID
   |clé=     Hal:inria-00100513
   |texte=   Signal transformation strategies to improve speech intelligibility for second language acquisition
}}