T3 Speech Recognition, powered by Natural Speech Communication (NSC), enables highly reliable speech recognition from telephones, cell phones and speaker phones. More intuitive than a touch tone interface, T3 Speech Recognition allows callers to use simple and intuitive voice commands to get immediate access to the information they need, decreasing operational costs and increasing customer satisfaction.
Global Directory - Using natural voice commands, callers no longer have to memorize extensions or know how to spell an individual's name. Through Address Book, internal callers can look up names directly from Microsoft®Exchange® or other address books. This feature may be password protected.
Voice Activated Messaging - Access voice messages through voice commands, eliminating the need for touch tone operation of voice mail. Imagine if you no longer had to memorize confusing voice mail codes and simply inquired to "play messages from yesterday" or "check if I have a message from Mr. Jones."
Interactive Voice Response - Optimize service center quality and operational costs through "voice-driven" services. Callers can easily access account information, register for events, or activate services when IVR is voice-enabled.
Speech Enabled PBX - Simplify usage of phone systems through T3-enabled speech recognition on PBX environments. Functions that normally consist of multiple steps, such as forwarding a phone extension, can be programmed to accept simple voice commands.
| Automatic Speech Recognition (ASR) engine | Incorporates phoneme-based recognition, Speaker Independent Recognition (SIR). |
|---|---|
| Compact hardware solution | Because the NSC speech board resides in the same server as T3main, additional CPU resources are not required. |
| Dynamic resource allocation | T3main dynamically manages speech board resources. Resources can be split when less processing power is needed, reducing costs. |
| Cost effective servers | The speech board provides the extra resources needed for speech recognition capabilities, allowing lower capacity servers to be used for smaller applications. |
| Automatic Speech Recognition (ASR) engine | Incorporates phoneme-based recognition, Speaker Independent Recognition (SIR). |
| High-end, high density capabilities | High density boards and up to 120 concurrent speech recognition channels per board allow the use of only one server even with a significant amount of voice traffic. |
| Barge-In | A context-sensitive filter helps reduce false barging due to background noise and non-relevant utterances. |
| Full DTMF | Touch tones are available to callers. |
| Internationalized System | Languages include English (US and UK), Spanish (US), Russian, German, Arabic and Hebrew. |