Opened 3 years ago

Last modified 2 years ago

#12819 new enhancement

Implement a speech recognition application

Reported by: richienyhus Owned by: nobody
Priority: normal Milestone: Unscheduled
Component: Applications Version:
Keywords: Cc:
Blocked By: Blocking:
Has a Patch: no Platform: All


Another form of accessibility assistance is system speech recognition.

Implementations on other systems include: Windows Speech Recognition, Apple Speech and Dictation, KDE Simon, Gnome Vendics.

Voice recognition engines that we can use:

Library Language Licence
CMU's PocketSphinx C BSD
CMU's Sphinx3 C BSD
Kaldi C++ Apache
Julius C Attribution Assurance

Change History (2)

comment:1 Changed 3 years ago by richienyhus

Windows and OSX both have a basic voice user interface (VUI) by using spoken commands called common commands on Windows and speakable items on OSX. There was an aborted effort on Gnome called Gnome VoiceControl, but Gnome Vendics can handle limited English voice commands.

Edit: there is also Voicely (MIT licenced) which could be forked and ported to Haiku, or used as reference for a C++ Haiku implementation.

Last edited 3 years ago by richienyhus (previous) (diff)

comment:2 Changed 2 years ago by mmu_man

For reference, my antique BeOS patch for CMU Sphinx2. Altough the native audio input seems to refer to a DataIOSoundRecorder.h that was probably added in ZETA...

Note: See TracTickets for help on using tickets.