|
|
SpeechRunner delivers complete testing of
voice-enabled applications with interactive test development, stand-aside
test execution, and automated results analysis. SpeechRunner supports SAPI
5 compliant voice-recognition engines as well as implementing a
simulated voice-recognition engine for highly reliable functional testing.
SpeechRunner presents a unique
side-by-side simulation of voice interactions with the target application,
letting you explore test possibilities as you compose high-level tests.
Unlike black-box test composers, SpeechRunner connects to the target
application much like a debugger: by stepping phrase-by-phrase through
your application, you can create high-level tests that will likely remain
usable through dramatic changes in your user interface.
SpeechRunner phrase management catalogs
phrases required for tests, driving collection and playback of recorded
voice files. Post-test analysis tools help you leverage collections of
voice samples to analyze, enhance and upgrade your voice user interface as
voice recognition technologies evolve.
Key Features of SpeechRunner
- Provides a familiar workspace
look-and-feel for Microsoft Visual Studio users.
- Tests any SpeechStudio enabled
application without modification.
- Promotes sharing component-based test
modules in multiple tests and applications.
- Side-by-side simulation allows the test
developer to step through the test to understand voice effects and edit
based on intelligence from the executing application.
- Cooperates with SpeechStudio to make
more intelligent choices available for test construction.
- Tests remain high level and easily
intelligible, maximizing their longevity despite changes to the test
application.
- Full simulation allows developers to
verify and debug applications and tests without the delays and
uncertainties of actual recognition.
- Uses standard phrase libraries to
leverage conversation databases,
decreasing the cost of building and maintaining test recordings.
- Context-sensitive help provides
assistance when you need it.
- Extensive online tutorial and reference
manual.
- Measures test success automatically
using feature-extracting differences to ignore variations such as timing
and voice recognition uncertainties.
- Synchronizes high-level commands,
allowing tests to remain flexible and efficient despite recognition
failures and wide variations in engine and host speeds.
- Implements control commands to set up
specific engine conditions to diagnose and assess engine performances.
- Maintains phrase-level histories
supporting voice interface refinements to avoid misrecognitions before
your application is launched, saving time, money and enhancing your end
user satisfaction.
- Automated regression testing allows your
application to evolve while catching errors as early as possible.
- Integration with other testing tools,
such as Rational Visual Test, preserves existing testing investments.
- High-level test commands are independent
of speech engines and recording formats. Voice recognition technology
can continue to evolve without undermining your test investment.
- Designed for use in a multi-program
environment, where multiple applications may cooperate or compete for
voice attention.
|
