The legacy V1 API that the Real-time Virtual Appliance currently supports will be discontinued in a future release. We recommend that customers familiarise themselves with the configuration object as described in the Speech API Guide. Future notices will be provided to announce the end of life of the V1 API, and provide detailed instructions on migrations to the V2 API.
Currently, Speechmatics supports 2 python libraries for use with our Real-time products. smwebsocket-py is recommended for use for the Real-time Virtual Appliance only, and speechmatics-python is recommended for use in both our Real-time Container and our Real-time Virtual Appliance. In a future release we will exclusively support speechmatics-python as our preferred Python library. We recommend you familiarise yourself with this library and use wherever possible. Please contact support@speechmatics.com if you require access to this library.
The following issues are addressed since the previous release:
| Issue ID | Summary | Resolution Description |
|---|---|---|
| REQ-15418 | Custom dictionary with splitting characters gets incorrect pronunciation | When using words with splitting characters in a Custom Dictionary (for example covid-19) where a number follows a word we now have the correct pronunciations created. Splitting characters include ["-", "_", "/", "<", ">", ":", " "]. This is for all languages |
| REQ-11426 | When there are two speakers in German, the second speaker is sometimes not transcribed | This is now resolved in the latest German release |
| REQ-11960 | Transcripts are blank for some German audio files | This is now resolved in the latest German release |
| REQ-12791 | Some German numbers are not transcribed accurately or are missing | This is now resolved in the latest German release |
| REQ-12792 | Some German transcripts are missing words | This has now been resolved in the latest German release |
| REQ-15870 | In the global English language pack the word "check" with British output locale is always transcribed as "cheque" | This is now fixed in the English language pack |
The following are known issues in this release:
| 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 ' |
| REQ-7549 | Memory leak affecting gRPC | There is a small memory leak in the gRPC Python server (https://github.com/grpc/grpc/issues/5913). |
| 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-11087 | Additional white space appearing before the very last end of sentence punctuation character. | Its been observed that additional white space can occur before a full-stop '.' in some transcripts. This can be parsed and removed by a post-processing step if necessary. |
| REQ-11135 | 3.2.0 introduced unwanted hesitations in transcripts. | Due to changes in the way that training data is now ingested to improve the accuracy of spontaneous speech for English (en) there is a greater likelihood that hesitations will be included in the output transcripts. We plan to support a hesitation filtering capability in a future release for customers that do not want to see hesitations on transcripts. |
| REQ-11792 | Speaker change token positioning is incorrect | We are aware of a consistent mis-placing of the speaker change |
| REQ-12202 | High memory usage when using custom dictionary | It has been observed that when using custom dictionary an additional 800-1700MB of memory is required (depending on the size of the wordlist used). |
| REQ-16256 | Heavy usage of RAM when swapping between 8kHz and 16kHz input | Where multiple persistent workers are configured with Custom Dictionary that swap between 8kHz and 16kHz input, this can cause a memory leak that causes the container to crash. If this is a workflow used it is recommended to consistently check the memory usage of the appliance, and restart it where necessary |
Virtual Appliance image (OVA) for installation on:
See the Installation and Admin Guide for details on the minimum specifications for the VM. The maximum number of concurrent jobs (maxworkers) that you can run on a single appliance is 30.
There are four variants of the Real-time Virtual Appliance.
| Variant | Image Size | Max. Disk Space | Languages |
|---|---|---|---|
| nano | 10GB | 40GB | en |
| mini | 14GB | 40GB | en, de, es |
| midi | 26GB | 60GB | en, de, es, fr, ko, ja, nl, pt |
| maxi | 42GB | 80GB | en, de, es, fr, ko, ja, nl, pt, it, da, pl, ca, hi, ru, sv |
| plus | 46GB | 80GB | en, cmn, no, ar, bg, cs, el, fi, hu, hr, lt, lv, ro, sk, sl, tr, ms |
Remove the license from your old appliance (see the Admin Guide), then re-import the new OVA and configure networking as per the Installation and Admin guide. You will need to re-apply the license code you have once the OVA has imported.
Upload the OVA to VMWare ESX, VMWare Workstation Player, or VirtualBox. See the Installation and Admin Guide for more information.