OLIVE 5.4.0 Release Plugins

The following plugins have been tested and certified for compatibility and release with the OLIVE 5.4.0 software package.

Speech

• Speech Activity Detection (SAD)
  • sad-dnnSmolive-v1.0.0 (low resource)
  • sad-dnn-v7.0.2
• Voice Type Discrimination (VTD)
  • vtd-dnn-v7.0.2
• Deep Fake Audio Detection (DFA)
  • dfa-spoofnet-v1.0.0
  • dfa-speakerSpecific-v1.0.0

Speaker

• Speaker Identification (SID)
  • sid-dplda-v2.0.1
  • sid-embedSmolive-v1.0.0 (lower resource)
• Speaker Detection (SDD)
  • sdd-diarizeEmbedSmolive-v1.0.2 (lower resource)
• Speaker Diarization (DIA)
  • dia-hybrid-v2.0.2
• Speaker Highlighting (SHL)
  • shl-sbcEmbed-v1.0.2

Language

• Language Identification (LID)
  • lid-hdplda-v1.0.1
  • lid-embedpldaSmolive-v1.0.0 (lower resource)
  • lid-embedplda-v3.0.1 (Legacy)
• Language Detection (LDD)
  • ldd-embedpldaSmolive-v1.0.0 (lower resource)
  • ldd-embedplda-v1.0.1

Gender

• Gender Identification (GID)
  • gid-gb-v2.0.1
• Gender Detection (GDD)
  • gdd-embedplda-v1.0.0

Keyword

• Query By Example (QBE)
  • qbe-ftdnnSmolive-v1.0.0 (lower resource)

Transcription

• Automatic Speech Recognition (ASR)
  • asr-dynapy-v3.0.0

Translation

Note that this TMT plugin currently has additional server and resource restrictions. Contact SRI for details.

• Text Machine Translation (TMT)
  • tmt-neural-v1.0.0

Topic

• Topic Detection (TPD)
  • tpd-dynapy-v5.0.1
  • tpd-fusion-v1.0.1

Manipulation

• Audio Redaction (RED)
  • red-transform-v1.0.0

Imagery

• Face Detection Image (FDI)
  • fdi-pyEmbed-v1.0.0
• Face Detection Video (FDV)
  • fdv-pyEmbed-v1.0.0
• Face Recognition Image (FRI)
  • fri-pyEmbed-v1.0.0
• Face Recognition Video (FRV)
  • frv-pyEmbed-v1.0.0

For further information on each individual plugin and their capabilities, please refer to the pages linked above. For more information about how to integrate with OLIVE and get started using these plugins, refer to the appropriate documentation linked from the landing page.