TI eXpressDSP compliant algorithm
TI eXpressDSP (XDAIS) Compliance Benefits
VoiceCom offers to the developers of DSP
applications a possibility to enable their products with
voice interface, which is a natural and convenient way
to control any device.
u Voice command (single-word/phrase) recognition
technology. Provides speaker-dependent
recognition (user trains each command before use)
and speaker-independent recognition for a limited
set of commands (user can use a set of commands
without previous training) which is done by
collecting a voice database for this set in advance.
u Low resource consumption. Even resource-limited
DSP-based devices (mobile equipment, PDAs,
etc) now can have powerful voice interface.
u Language-independence. The technology can be
used for any language, while most competitors'
technologies are limited to a few languages.
u Noise robustness. VoiceCom technology includes
low resource consuming noise cancellation. For
applications in a noisy environment, VoiceCom can
be combined with Denoiser Clear Voice, a powerful
noise cancellation technology.
And the most important feature of VoiceCom is the
simplicity of development. Any product can be quickly
and efficiently enabled with such an advanced option
as Voice Interface.
Possible Applications :
u Consumer electronics - Use Voice commands
rather than IR remote controls.
u Industrial Equipment - Many industrial
applications where operators need to have their
hands free for other tasks.
u Hands-free sets - drive and speak on your mobile
safely, because all you need to dial a number or to
search the phonebook is your own voice.
u PDAs - If it is not possible to design a portable
device with a keyboard, just make it controlled by
voice - it's easy.
In general, VoiceCom allows to control by voice
almost anything and makes life easier.
VoiceCom is fully TI eXpressDSP compliant
algorithm and this guarantees time and money
saving in building in any DSP solution.
This algorithm follows the run-time conventions
imposed by TI's implementation of the C prog
This algorithm is re-entrant within a preemptive
environment (including time-sliced preemption).
All algorithm data references are fully
All algorithm code is fully relocatable.
This algorithm does not directly access any
This algorithm does not include definitions
specific to a debug variant.
- Code 20 KW
- Data 10 KW
Processor requirements 2.5 MIPS
By request, the support of other platforms may be
developed by experienced and professional DSP
engineers of Speech Technology Center,
customization is possible according to the
Document Path: ["1153-speech-technology-center-brochure-voicecom.pdf"]