Proprietary tools like Dragon Naturally Speaking & Sonic Extractor are good options. Some Open Source softwares like SHoUT & Sphinx are also available, but their Voice pattern recognition engine & performance may not be on par with the proprietary ones.