| 摘要 |
<p>The invention concerns a speech detection method in an audio signal comprising a step which consists in obtaining an energy information of the audio signal, said energy information being used to detect speech in the audio signal. The invention is characterized in that the method further comprises a step which consists in obtaining a vocalizing information of the audio signal, said vocalizing information being used jointly with the energy information for speech detection in the audio signal.</p> |
