Following laryngectomy only 40 per cent of patients are able to attain esophageal speech. Several surgical and nonsurgical methods have been devised to solve this problem, but non is fully satisfactory. This paper describes two different experimental apporaches to the problem. A segment of trachea is isolated with an intact blood supply and anastomosed to the skin and esophagus, thus producing a fistula where the characteristic of the tracheo-esophageal junction can be studied. Efforts to implant a valve into this fistula have been unsuccessful but the fistula alone is patent after one year. The isolated tracheal segment provides an ideal model for study in an untracheatomized animal. In the second approach an electromagnetic sound device is being devised which can be implanted in the neck as near as possible to the previous site of the larynx. The results have been encouraging.