HEAD Acoustics Launches Voice Control Analysis Software
German speech recognition software provider HEAD Acoustics has released VoCAS (Voice Control Analysis System), a software package for evaluating speech recognition systems.
From voice control in vehicles to speech commands for smartphones, tablets, or telephone hotlines, VoCAS allows quick and objective quality evaluation of automatic speech recognition systems under realistic and reproducible test conditions. The software takes crucial factors into account, such as background noise, language, or accent, which can significantly influence the performance of voice control systems. VoCAS uses predefined test sequences for ASR systems to determine their quality, to analyze the weaknesses of the systems, and to optimize them based on the results.
Depending on the device under test (DUT) and the requested test case, the appropriate test sequence can be defined in VoCAS. Each test sequence consists of different elements and is processed sequentially. These elements are, for example, background noises, inserting pauses for acoustical feedback of the voice control system, or the evaluation of the DUT. All elements can be arranged, added, and adjusted individually as needed. The user can also choose various parameter sets, such as different speakers, languages, background noises, destination addresses, or people to be called.
For testing different voice control systems, audio source databases with appropriate speech commands are required. Audio databases in VoCAS can be individually expanded by importing speech recordings. In addition, VoCAS provides an integrated recorder for recording individual speech commands. Larger lists of imported or recorded audio files can be cut, filtered, and adjusted to defined speech levels automatically. VoCAS also allows users to manually tag keywords to each speech command and guides the user through the requested variants, creates the appropriate measurement sequences, and helps to keep an overview.
VoCAS also offers a clear representation of test results as percentage values (60 percent of speech commands recognized, 40 percent not recognized) or colorful accentuation. Furthermore, a direct comparison of voice control systems is possible. Results can be exported to Microsoft Excel for further post processing.
VoCAS is compatible with other HEAD acoustics products. The front-end MFE VI.1 can be controlled via VoCAS for playback and monitoring of speech recordings via the artificial head measurement system and for mouth equalization. Furthermore, the background noise simulation systems 3PASS, HAE-BGN, or HAE-car can be managed via VoCAS.