Skip to main content
Skip table of contents

VDK Runtime 6.1.0 - 14th January 2026

Summary

We’re excited to announce the release of VDK Runtime 6.1.0, introducing VEC (Voice Error Correction) — a new post-processing module designed to significantly improve Automatic Speech Recognition accuracy for error-prone inputs such as alphanumeric sequences. This release enhances reliability for voice-driven workflows without requiring any changes to existing application code.

What's New in VDK Runtime 6.1.0?

Voice Error Correction (VEC)

VDK 6.1.0 introduces VEC, a lightweight and high-performance module that improves ASR output quality by correcting common recognition errors after decoding.

VEC can leverage a context list — a predefined set of valid alphanumeric sequences — to further improve accuracy. By favoring corrections that match known valid entries (such as product codes or customer IDs), VEC reduces semantically invalid results even when ASR output is phonetically plausible.

Learn More

Additional updates and refinements have been made across the VDK ecosystem to enhance developer experience and performance.

We’re committed to improving usability, flexibility, and performance — and your feedback on VDK 6.1.0 is always welcome.

Versions

Samples (C++)

Component

Package name

Version

Technologies

Chained grammars

chained-grammars

6.3.0 UPDATED

VOICE RECOGNITION (GRAMMAR)

Dynamic grammar

dynamic-grammar

6.3.0 UPDATED

VOICE RECOGNITION (GRAMMAR)

Simple application

simple-application

6.3.0 UPDATED

VOICE RECOGNITION (GRAMMAR) VOICE SYNTHESIS

Tts

tts

6.3.0 UPDATED

VOICE SYNTHESIS

Voice biometrics

voice-biometrics

6.3.0 UPDATED

VOICE BIOMETRICS

Voice Commands Language Understanding

voice-commands-language-understanding

3.3.0 UPDATED

VOICE RECOGNITION (CONTINUOUS)

Speech Enhancement

speech-enhancement

2.3.0 UPDATED

VOICE RECOGNITION (GRAMMAR) SPEECH ENHANCEMENT

VSDK (C++)

Component

Version

Vsdk

10.4.0 UPDATED

Vsdk-csdk-asr

1.5.0 UPDATED

Vsdk-vec-addon

1.1.1 NEW

Vsdk-csdk-tts

1.5.0 UPDATED

Vsdk-idvoice

5.3.0 UPDATED

Vsdk-tssv

4.3.0 UPDATED

Vsdk-vnlu

3.3.0 UPDATED

Vsdk-s2c

2.3.1 UPDATED

Vsdk-audio-portaudio

4.4.0 UPDATED

Samples (JAVA)

Component

Package name

Version

Technologies

Chained grammars

chained-grammars

7.1.0 UPDATED

VOICE RECOGNITION (GRAMMAR)

Dynamic grammar

dynamic-grammar

6.1.0 UPDATED

VOICE RECOGNITION (GRAMMAR)

Free speech

free-speech

4.0.0

VOICE RECOGNITION (CONTINUOUS)

Simple application

simple-application

6.1.0 UPDATED

VOICE RECOGNITION (GRAMMAR) VOICE SYNTHESIS

Tts

tts

7.1.0 UPDATED

VOICE SYNTHESIS

Voice biometrics

voice-biometrics

4.1.0 UPDATED

VOICE BIOMETRICS

Voice Commands Language Understanding

voice-commands-language-understanding

1.0.1

VOICE RECOGNITION (CONTINUOUS)

Speech Enhancement

speech-enhancement

3.1.0 UPDATED

VOICE RECOGNITION (GRAMMAR) SPEECH ENHANCEMENT

Vdk service

vdk-android-sample-service

1.0.0

VSDK (JAVA)

Component

Version

Vsdk

7.2.0 UPDATED

Vsdk-csdk-asr

3.2.0 UPDATED

Vsdk-vec-addon

1.1.0 NEW

Vsdk-csdk-tts

3.2.0 UPDATED

Vsdk-idvoice

4.2.0 UPDATED

Vsdk-tssv

4.2.0 UPDATED

Vsdk-vnlu

1.1.0

Vsdk-s2c

3.2.1 UPDATED

Details

VSDK

Vsdk

Features:
  • Voice Error Correction (VEC)

    • New ASR post-processing module for correcting alphanumeric recognition errors

    • Structure-preserving: ASR result schema remains unchanged

    • Supports 1–7 character sequences (lowercase letters and digits)

    • Language-dependent; requires matching acoustic model and lexicon

    • Optional predefined list of valid sequences

    • Supports very large lists (hundreds of thousands of entries)

    • <10 ms processing latency per recognition

JavaScript errors detected

Please note, these errors can depend on your browser setup.

If this problem persists, please contact our support.