Batch Container

Important Notices

It is now necessary to use processors that support Advanced Vector Extensions 2 (AVX2) when running the container in all scenarios in order to take advantage of latest performance optimisations.

It is also recommended when using the enhanced model to use hardware that supports the AVX512_VNNI flag for optimal processing performance. The enhanced model also has increased compute requirements and will run more slowly than the standard model. For more information please see the quick start guide.

9.1.0

New

• New English finance domain language pack. Provides accuracy improvements when specific financial jargon is spoken in your audio. Refer to documentation here for more details
• New language Ukrainian (uk)
• 16 Languages updated with additional punctuation marks for improved readability
  • The following languages now support (. ? , !): Bulgarian, Catalan, Czech, Greek, Finnish, Croatian, Hungarian, Lithuanian, Latvian, Norwegian, Polish, Romanian, Slovak, Slovenian, Ukrainian, Korean
• New parameter added for controlling Speaker Diarization sensitivity: speaker_sensitivity. Refer to our documentation here for more details

Improved

• Improved accuracy for French, including more data for Canadian French (fr-ca)
• Improved accuracy for Portuguese, including more data for Brazilian Portuguese (pt-br)
• Improved accuracy in standard operating point for Romanian, Hungarian, Danish, Slovakian, Croatian, Bulgarian, Finnish, Slovenian, Lithuanian
• Updated Danish, Norwegian and Swedish to remove undesired character sets
• Improved accuracy in localised spelling for English output locale feature
• Improved accuracy of percentage symbol recognition in French
• Speaker Diarization can now utilize multiple cores in parallel, significantly increasing transcription speed and RTF. More information about parallel processing can be found here

Fixed

• Fixes for English and Italian written form numeric entities
• Fix for handling small number of files with multiple audio channels were mistakenly detected as containing inverted audio, which lead to no transcription being returned
• Resolved an issue where auth headers for the fetch URL feature were not sent correctly

Known Limitations

Issue ID	Summary	Detailed Description and Possible Workarounds
REQ-1409	Proteus HCL with <unk> causes out of memory error	A custom dictionary list that contains the word '' causes the worker to crash.
REQ-10160	Advanced punctuation for Spanish (es) does not contain inverted marks.	Inverted marks [ ¿ ¡ ] are not currently available for Spanish advanced punctuation.
REQ-10627	Double full stops when acronym is at the end of the sentence	If there is an acronym at the end of the sentence, then a double full stop will be output, for example: "team G.B.."
REQ-10634	Putting "-" as an item in additional vocab configuration will cause the container to fail	Do not enter just a "-" on its own in Custom Dictionary either as an additional vocab item or in the sounds_like property. Hyphens are still supported when entered as part of phrases or words

Supported Platforms

Docker (17.06.0+) running on Ubuntu, Debian, Fedora or CentOS

Installation

Pull the Batch Container Docker image from the Speechmatics Docker repository

Pre-requisites

You have a login (URL, username and password) for the Speechmatics Docker repository, and have a Docker environment (version 17.06.0 or above) running

Supported Languages

Below is the complete list of languages supported by Speechmatics.

Speechmatics takes a global first approach to our languages. In a single language pack we aim to support many different accents and dialects. This simplifies your workflow when selecting which language to use, not requiring you to know which accent is being spoken in your audio up-front. With this approach we still achieve very high accuracy compared to accent specific language packs.

Language	ISO Code
Arabic	ar
Bulgarian	bg
Catalan	ca
Mandarin	cmn
Czech	cs
Danish	da
German	de
Greek	el
Global English	en
Global Spanish	es
Finnish	fi
French	fr
Hindi	hi
Croatian	hr
Hungarian	hu
Indonesian	id
Italian	it
Japanese	ja
Korean	ko
Lithuanian	lt
Latvian	lv
Malay	ms
Dutch	nl
Norwegian	no
Polish	pl
Portuguese	pt
Romanian	ro
Russian	ru
Slovakian	sk
Slovenian	sl
Swedish	sv
Turkish	tr
Ukrainian	uk
Cantonese	yue

Container images are labelled using the following scheme, where language codes adhere the ISO-639 standard:

batch-asr-transcriber-<language>:<version>

For example,

batch-asr-transcriber-en:9.1.0