Done, set up a MIT license in the repo.
Regarding the accuracy, it really depends on how clear the caller’s voice is. If there’s a lot of noise in the background it can have trouble. I find it works pretty well for my needs.
I will definitely give it a try.
Implemented it this today. It works great except that the Bing Speech to Text REST API only allows for up to 15 seconds of audio. It works for the short voice mails but obviously not the longer ones.
Anyone willing to make this a standard module?
We used this script for a month, didn’t have issues with messages longer than 15 seconds. The script work very well, thank you.
The issue we had, Bing sucks. The transcriptions were terrible, script was great.
We ended up transitioning to Watson, used some so Josh’s techniques along with some other scripts we found.
Would love to see both services offered as a module, I don’t have the skill set to create one.
Would you mind sharing the Watson version of this?
There is an inelegant but working script here for IBM Bluemix:
Does this involve creating a BlueMix account and then creating the API in BlueMix?
Nevermind I think I found it under the Watson api commands
It would be awesome if we could add transcriptions to UCP so we can provide HIPPA compliant voicemail transcription.
With UCP 13. You could add a new field to the VM and put the transcript into the text file which would show up in UCP. Getting it from the script to that text file is the part I stopped on. Basically text to file though.
This is really great!
I implemented this and am now wondering, anyone have any luck getting it to appear more “grammatically correct”? The text I have returning is one long run-on sentence.
Supposedly, Watson should be able to place punctuation, grammer, etc.
You meant HIPAA. And, yes…agree that transcriptions would be a necessary feature to add.
Hmmm…Set this up because a Customer wanted it - the Accuracy through Bing is actually pretty spot-on, but it stops transcribing after the first pause - which is a No-Go.
Cognitive Services Bing Speech API is being retired November 1, 2021 so it won’t work after then anyways - Has anybody looked at translating this to the new Azure Speech Service - I am trying right now, but I have almost Zero experience with Python - If I get it translated I will post it here - the Microsoft docs say it’s easy, but it is certainly not easy for me. Will update as I progress (or don’t) this weekend.
gcloud STT works for me, they have bindings documented for most likely languages
Ok - Gave up on Microsoft - I am just too weak with Python!
Works perfectly and getting set up with IBM is trivial - and right now they are giving out a $200 credit when you set up a real account (not a free one).
Why isn’t this a standard thing in FreePBX - You could leave the script like it is without any proper credentials, so that it just wouldn’t work unless you had the credentials, but DANG - it works perfectly and the recognition is really spot-on!
This is so cool and really up’s the “Polish” level of FreePBX - Lots of Hosted providers are providing Transcription - now we can too!
FreePBX Weather Forecast
That specific script is built on work with no explicit license terms. I’ve noted that in the notes at the top, and presumably anyone who is using it is comfortable with that. But it does make it unsuitable for inclusion in the freePBX core.
One of the goals of Asterisk 19 (due out in the fall) is improved support for TTS and STT. Once we see that take shape, ‘proper’ FreePBX support will follow.
Great - Like I said in another post, I am getting back to my On-Premise boxes (we have been focused on Hosted for years now) and this is one of the key things that hosted had an exclusive on that is no longer the case.
I am also experimenting with your Web Phone - assuming that works the way it’s advertised, FreePBX is a real option when compared to Hosted from other vendors - and trunking to a FreePBX is way cheaper than hosted especially on larger installs (30+ Phones).
Great work - Thanks!
This topic was automatically closed 31 days after the last reply. New replies are no longer allowed.