Amazon Web Services (AWS) Transcribe Speech Recognition Plugin 1.5.0 to the UniMRCP Server (UMS) has been released.
The plugin is based on the following components:
- UniMRCP Server 1.7.0
- AWS SDK 1.9.181
- Libevent 2.1.9
The binaries are currently available for the following Linux distributions:
- Red Hat / CentOS 7 (unimrcp-transcribe-1.7.7-1.el7.x86_64.rpm)
- Red Hat / CentOS 8 (unimrcp-transcribe-1.7.7-1.el8.x86_64.rpm)
- Ubuntu 18.04 LTS (unimrcp-transcribe_1.7.7-bionic_amd64.deb)
- Ubuntu 20.04 LTS (unimrcp-transcribe_1.7.7-focal_amd64.deb)
This release adds support for language identification and provides a few fixes and improvements in the existing functionality. The release is built against a newer version of AWS SDK.
The detailed list of changes introduced in this release follows.
New Features
- Populate and set confidence score in NLSML results.
- Added support for language identification.
- Allow 'logging-tag' to be specified via vendor-specific parameters.
- Added support for sorting result alternatives based on confidence score (the highest first).
Fixed Problems
- Fixed a possible segfault on processing the header field Logging-Tag having an empty [no] value.
- Fixed support for HTTP proxy.
Configuration Parameters
- Added new attributes 'identify-language', 'language-options', 'preferred-language' to the element 'streaming-recognition'.
- Added new attribute 'sort-alternatives' to the element 'streaming-recognition'.
Miscellaneous
- Upgraded the AWS SDK from 1.8.129 to 1.9.181 version.
- Added a new log statement on creation of an AWS profile.
- Updated the Usage Guide to reflect the changes introduced in this release.
Visit the Transcribe plugin page for more information.
http://www.unimrcp.org/transcribe
Thank you for using UniMRCP.
--
Arsen Chaloyan
Author of UniMRCP
http://www.unimrcp.org