Speech to text conversion by reading a wave file using stream RRS feed

  • Question

  • Hi,

    If we convert speech to text using a stream in the following manner-

    recEngine - SpeechRecognitionEngine, and
    wavStream - a simple System.IO.Stream which is actually a stream to an existing wave file

    My query is that "Reading of the wav stream starts from the header of the wav file, or is it that the header is skipped internally and directly the data is read?"

    If I set the wavStream.position to some value say "165", and suppose the header of the wav file was of size 120 bytes (just suppose, I am considering an example) then the speech to text will always give an error in case the stream is read starting from the data (skipping the header)..

    Any help highly appreciated.

    Thanks in advance,
    Wednesday, December 23, 2009 12:43 PM