SpeechToText
Jump to navigation
Jump to search
Speech to Text
[edit | edit source]This is a page with resources, planning, and progress for taking audio and outputting text. Even a not great version of this technique has multiple applications, such as extracting keywords from conversations.
Software Frameworks
[edit | edit source]- CMU Sphinx http://cmusphinx.sourceforge.net/html/cmusphinx.php
- Vista Speech API
- Julius http://julius.sourceforge.jp/en_index.php
- WaveToText (Windows shareware, uses MS Speech API) http://www.research-lab.com/vexp007read.htm
Training Corpuses, Resources
[edit | edit source]- http://www.voxforge.org/
- grey's srt idea for movies (basically - use SRT files or Closed Captioning as "automated" training to audio from films)