PORTEFEUILLE

Laryngectomy denotes a treatment of laryngeal cancer, which may require partial or total removal of the larynx. The loss of the vocal function is a distressing consequences of this pathology. Medical rehabilitation, such as oesophageal voice reproduction, surgical techniques and device techniques, allow laryngectomees to recover partially the vocal function. However, all techniques for voice rehabilitation result in poor voice quality and insufficient speech intensity. The LARYNX feasibility focuses on a market study and a preliminary concept of a vocal aid for laryngectomees, which improves speech quality through perceptual signal processing and increases speech intensity.

The project requires multidisciplinary scientific and technical competencies in different domains, such as speech processing and speech synthesis, low-power embedded systems and the medical expertise in the domain of the laryngectomy. Therefore, the project involves three research institutes with a proven experience in the biomedical field and in the development of innovative speech processing algorithms. The project consortium CSEM, EiVd and CHUV-ORL ensure the required merge of the medical expertise of voice rehabilitation and the engineering aspects of speech processing and voice reconstruction.

The proposed approach is exclusively based on a non-intrusive speech restoration by means of a digital signal processing unit.
The performed Market and Technology Study confirms the clear need for a non-invasive method, i.e. without any auxiliary surgery, requiring only modest production effort and reduced adaptation period from the patient. Worldwide 45’000 new larynx cases are expected yearly and an estimated 30% of these new larynx cases, which do not manage to achieve successful voice restoration after surgery through one of the three traditional methods, are left behind.
The results obtained during a Feasibility Study show a significant improvement of the quality of the restored voices and clearly demonstrate the feasibility of the proposed approach. The method was tested on a database of phonetically equilibrated and voiced speech obtained after supracricoid partial laryngectomy. The results show a considerable improvement of the voiced phonation characteristics. The improvement is particularly important for female partial laryngectomee. In the future, this technology could be adapted to voice improvement of total laryngectomee and also to very damaged voice of any etiology.
The results have been presented at different conferences with great successs. In order to ensure the continuation of the activity, contacts with the “Schweizer Krebsliga” are on-going.

“LARYNX – vocal aid feasibility for laryngectomees”, Josep Sola, Rolf Vetter, Philippe Renevey, Valerie Schweitzer, CSEM Technical Report, Jan. 2005 (deliverable)

keine