iSoftstone iSoftStone Work

[Neboz]

got the email to start, just read through the info a couple of times and not sure I want to think of 10 prompts for each scenario! Would be easy if they provided the utterances and let us do it on iPhone again!

 

[Savvy cash]

All being well with the recordings, I've just finished the task. It took me about an hour and a half to do in total, the hardest bit was just thinking of alternative phrases to say and coming up with the 10 different examples. I wrote all mine down and numbered them to speed things up when saving the files etc.

If you persevere it's not too bad when you get going with the recording software etc, just a bit mind numbing with saying "hey mercedes" over and over again.......

 

[cnorton]

Hi,
I receive emails from isoftstone to sign their agreements, I do so, and then I never hear back from them. On my profile it says:

2019-1671	Speech recognition command creator - American English	New Submissions	12/29/2019	Withdraw
2014-1297	English Transcriber - UK	In Training	07/25/2019	Withdraw
2015-1406	Online Web Evaluator - United Kingdom	New Submissions	07/25/2019	Withdraw

But I don't know how to actually do any tasks for them. Any help would be gratefully received. Thanks.

 

[hanna.obrien.79]

Hi,
I receive emails from isoftstone to sign their agreements, I do so, and then I never hear back from them. On my profile it says:

2019-1671	Speech recognition command creator - American English	New Submissions	12/29/2019	Withdraw
2014-1297	English Transcriber - UK	In Training	07/25/2019	Withdraw
2015-1406	Online Web Evaluator - United Kingdom	New Submissions	07/25/2019	Withdraw

But I don't know how to actually do any tasks for them. Any help would be gratefully received. Thanks.

I have been in training for that transcriber job too although haven't done any sort of training. I have several jobs I have applied for just sitting waiting too.

 

[cnorton]

Good to know I'm not the only one. I wonder what's happening here.

 

[UKHope]

For those doing the hey mercedes one, or anyone with experience with this type of thing, when recording and it transcribes incorrectly, "seem" instead of "scene" for example, do you record it again or leave it and move on?

 

[Savvy cash]

For those doing the hey mercedes one, or anyone with experience with this type of thing, when recording and it transcribes incorrectly, "seem" instead of "scene" for example, do you record it again or leave it and move on?

I tried saying the same phrase about 3 times, each time it transcribed "seem," instead of "scene," so I just explained this in my email reply to isoftstone saying I'd completed the task.
I figured that they will have my other recordings anyway so they'll know what the original phrase was supposed to be!
I'm just waiting to see if they get back to me with any feedback.

 

[Lakadema]

For those doing the hey mercedes one, or anyone with experience with this type of thing, when recording and it transcribes incorrectly, "seem" instead of "scene" for example, do you record it again or leave it and move on?

I had the same problem. In fact, I had difficulty getting it to recognise and record my prompts. Either I mumble or my microphone is crap. This resulted in me making several attempts really trying to enunciate my words until I got it right! Very tedious and i couldn't see a way to delete the multiple incorrect transcriptions so I just left them as they were. There were a couple at the beginning which did not record the whole sentence but as I couldn't go back and change them. I left them.
I emailed the company to ask advice but I guess they wont reply until Monday which would be too late so I just submitted it as it was and hope for the best.

 

[UKHope]

Thanks Savvy cash
Thanks Lakadema
Over 10 attempts on "scene", so I think I'll just continue with it and submit it as "seem".
Yep can't find a way to delete I don't think there is.

 

[UKHope]

Thanks Jon for the link for job

 

[Thesavvyqueen]

Well done on completing the task! I have mine all written down and numbered ready. Was just wondering if you saved each phrase you said individually, resulting in having to send 70 files over to them? Or if they are expecting the 10 phrases in one file so that totals 7 files?
I hope that makes sense, I’m finding the instructions a tad vague to understand.

 

[UKHope]

I sent 70 files, that was my understanding of the instructions anyway.

Just rechecked instruction I'm 99% sure it's for each utterance.

 

[JenC10]

Did anyone else have any issues with the Mercedes bot not picking up any of the utterances?? The voice recorder app is picking up everything fine but the actual bot itself is picking up nothing.

 

[Scratch]

I've just finished. Pretty easy really, it took me 35 minutes to record the utterances and probably another 15 to figure out what needed to be done, rename the files, send them etc. Hopefully I did it right and it's accepted.

 

[Lakadema]

Did anyone else have any issues with the Mercedes bot not picking up any of the utterances?? The voice recorder app is picking up everything fine but the actual bot itself is picking up nothing.

I had this problem at first and found it was because the browser was blocking the microphone access. After a bit of trial and error, I managed to get it working.

 

[Lakadema]

Well done on completing the task! I have mine all written down and numbered ready. Was just wondering if you saved each phrase you said individually, resulting in having to send 70 files over to them? Or if they are expecting the 10 phrases in one file so that totals 7 files?
I hope that makes sense, I’m finding the instructions a tad vague to understand.

I started off thinking it was 7 files containing 10 utterances, one for each scenario, but in the end just did 70 in one file. The instructions were very unclear as they referred to the utterances as prompts in one of the instructions which confused me. It took me longer than 1 hour too.

 

[Thesavvyqueen]

I started off thinking it was 7 files containing 10 utterances, one for each scenario, but in the end just did 70 in one file. The instructions were very unclear as they referred to the utterances as prompts in one of the instructions which confused me. It took me longer than 1 hour too.

Yes, I thought it best to do 70 files in the end to be on the safe side. I found it took a long time to actually think and write down all the phrases, but it helped a lot when it came to recording them. Overall I think it took about 2 hours of my time, but I guess still worth it for the 40USD. I had an email last night from wetransfer to say my files had been downloaded but no further communication since, hoping all is good!

 

[hanna.obrien.79]

I have heard nothing back about this job either! Beginning to take it personally

 

[fee123]

I have heard nothing back about this job either! Beginning to take it personally

Don't worry, they don't like me either

 

[Neboz]

I should of went with my 1st instinct and never started this, it’s turning into a nightmare! It’s taking me about 10 attempts for the chatbot to pick up what I am saying correctly! I have only recorded 8 so far and have emailed her about it, not gonna bother with anymore unless I get an email back! Maybe I just have a rubbish mic! Edit: had an email back not to worry much about what the chatbot transcribes ...phew!