Features

Tuning uses prompts, grammars, call flow, and caller data to improve the speech application performance. The LumenVox Speech Tuner also provides the following features:

• Database Loading — Faster, more user–friendly database loading
• Sophisticated filtering tool — Gives users the power to pinpoint specific calls or those matching certain user–defined criteria
• Updated Call Browser — Quickly grants users the ability to choose and listen to a specific interaction and export the audio file all from one window
• Improved statistics, Control Panel Interface and Shortcuts — Makes the transcription process easier
• Revised Grammar Tester — Allows users to instantly see how proposed changes to grammar or Speech Engine configurations will affect the results of the speech application

Tuning in 3 Easy Steps

Step 1: Import Data

Converters import call log data into the Speech Tuner database. All information stored by the call log is available in the Speech Tuner.

Step 2: Transcribe Speech

Transcribers can type the text of the caller's speech directly into the Transcriber. Once the audio is transcribed, the Speech Tuner compares audio transcripts with the Speech Engine results to determine accuracy, greatly reducing errors associated with manual evaluations. The transcripts are evaluated using the actual decode grammar, producing measurements such as word–error–rate, in– and out–of–grammar rates and semantic error rates.

Step 3: Test Immediately

Selecting an interaction in the Call Log automatically loads the associated audio and grammar into the Tester. The grammar can be edited, Speech Engine parameters set, and individual recognition tests generated. The Speech Tuner natively supports industry standard SRGS grammars. Once a set of possible changes is identified, users can batch test audio to evaluate performance, using those changes.

Call Browser

The Call Browser grants users the ability to quickly choose and listen to a specific interaction, and export the audio file all from one window. The Call Browser displays a list of currently loaded and filtered calls and displays all the interactions for a selected call.

The Browser window is divided into three sections:

• Calls List
• Interactions List
• Audio Control

Call List

Within the Calls List, you can easily move between calls to highlight a specific interaction. The data fields available provide key information such as call time, the number of interactions within a call, the number of times the Speech Engine recognized speech, and the confidence for times the Speech Engine correctly interpreted a phrase.

Interactions List

The Interactions List contains details of every interaction within a call. For each interaction, you can click the View Details button for a specific list of details including the acoustic model, decode time, NBest rank, and semantic interpretation.

Audio Controls

The Audio Control panel allows users to choose between hearing the decoded audio or the actual caller utterance with easy volume controls. An "Export as WAV" button provides users with the ability to export the call audio as a WAV file.

Tester

The Speech Tuner Tester makes changing grammars or other Speech Engine parameters secure, faster, and more accurate. The Speech Tuner allows you to use historical data to validate changes, ensuring the creation of a successful application.

Most tuning tools passively log views requiring that changes are made in the live speech recognition application requiring retesting over a period of time with live callers.

The LumenVox Speech Tuner is able to send changes back through the Speech Engine, simulating the recognition process and evaluating changes instantly. Instead of slow, non–interactive, static tuning, the Speech Tuner enables on–the–fly, highly interactive, dynamic tuning.

Make a change, do the test, get the results.

Grammar Evaluation

Speech can be evaluated against one or more grammar files similar to how the Speech Engine does in a production speech recognition system. Using the Speech Tuner's grammar editing tools, a grammar can be adjusted and re–tested to review if changes improve performance. This allows instantaneous grammar changes to improve speech recognition accuracy.

Performance Rate

The Speech Tuner rates performance against commonly accepted measures like WER (Word Error Rate). This helps give an accurate picture of details like average confidence scores, correct versus incorrect responses and in-grammar versus out–of–grammar performance.

Performance Rate also provides detailed information about an application's performance confidence scoring making understanding confidence scores easier. Simple graphs display the range of confidence scores for correct, incorrect, and out–of–grammar responses. The Speech Tuner will even calculate optimal confidence thresholds for a specific menu or application, taking the guesswork out of setting thresholds.

Parameter Evaluation

Changing Speech Engine parameters optimizes the Speech Engine performance, further improving the caller experience. Traditionally, changing Engine parameters has been a difficult and time–consuming task, often requiring long delays between changing a parameter and evaluating its effect on performance. The LumenVox Speech Tuner shortens this delay by offering dynamic testing capability. Speech Engine parameters such as search optimizations, speech end–pointing, and NBest result processing can easily be adjusted and immediately re–tested and re–scored from within the testing component.

Transcriber

Good transcripts are an important part of properly tuning a speech recognition application. The LumenVox Transcriber is designed to make this process quick and seamless.

In the most recent version of our Speech Tuner the transcription process is significantly easier with improved statistics, a new control panel interface and shortcuts.

Transcribing Audio

Transcribing speech is an excellent way to analyze how callers interact with the system.

Transcriptions are used to calculate automatic performance measurements such as in–grammar or out–of–grammar rates and recognition accuracy. Good transcripts enable the Speech Tuner to adjust the speech recognition application as needed.

Transcription Entry

The Transcriber is used to write down every word in a call. The Grammar Tester uses these transcripts in evaluating how well a Speech Engine is interpreting what users are saying.

Transcription Log

The Transcription Log allows detailed viewing of all transcripts during a transcription session. The Transcription Log tracks Interaction number, name and transcription description.

What's New:

Speech Tuner Version 10.1

Expanded Speech Tuner Tools — A counterpart to our new Text–to–Speech Engine (TTS), the LumenVox Speech Tuner 10.1 now allows users to fine–tune LumenVox TTS settings, view and edit SSML documents, preview TTS output. The new upgrade also sports an auto–complete system for speedier GrXML editing.