Hi, I am looking for a solution that can convert speech to text. Our idea is
to create an online portal where our customers can upload a audio file (wav,
mp3, wma, etc) and then take that file and convert the audio to text and
display it. Is or has sphinx done this? If so, any proof of concepts?
Is anyone out willing to quote us on the development for this project? We
would just need development work for the backend converting. We will make the
upload interface and display mechanism.
Thanks,
Chris
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi, I am looking for a solution that can convert speech to text. Our idea is
to create an online portal where our customers can upload a audio file (wav,
mp3, wma, etc) and then take that file and convert the audio to text and
display it. Is or has sphinx done this? If so, any proof of concepts?
Such solution is usually built with highly accurate automated transciption
engine and following human-guided text correction. Automated solution doesn't
provide enough accuracy. As a proof of concept you can check
Is anyone out willing to quote us on the development for this project? We
would just need development work for the backend converting. We will make the
upload interface and display mechanism.
$500k should be enough to launch this.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
It depends on your task and target application/audience. For machine
transcription to work (to be suitable for human view), it needs quite some
requirement to the audio itself (very little noise, speaker accent, etc).
If this is not possible, you'd have to use human, as nick suggested. There are
tons of human transcription companies out there. They typically offer at a
price of $2~3/minute, which is hefty if you are targeting regular people.
Alternatively, you can check out us, http://nexiwave.com . We offer speech indexing, which
you can let your user to quite accurately locate to the exact location in
their audio, through search. They can then request for a human transcription
if desired. We offer a Web Service API, so the backend has been taken care of.
You only need to worry about the frontend (and integration to human
transcription).
Ben
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi guys,
I do not know have you solved your problem or not but I wanna recommend a way of free audio to text converter https://audext.com/ You can choose any audio format you have and convert to text it during a few minutes. Hope, I was useful for you in this advice.
Last edit: Alona Serga 2019-09-26
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi, I am looking for a solution that can convert speech to text. Our idea is
to create an online portal where our customers can upload a audio file (wav,
mp3, wma, etc) and then take that file and convert the audio to text and
display it. Is or has sphinx done this? If so, any proof of concepts?
Is anyone out willing to quote us on the development for this project? We
would just need development work for the backend converting. We will make the
upload interface and display mechanism.
Thanks,
Chris
Such solution is usually built with highly accurate automated transciption
engine and following human-guided text correction. Automated solution doesn't
provide enough accuracy. As a proof of concept you can check
http://www.worldtech24.com/business/speakertext-automates-and-crowdsources-
video-transcripts-100-beta-invites
it runs using CMUSphinx. Another example is
http://www.3playmedia.com/
$500k should be enough to launch this.
Hi, Chris,
It depends on your task and target application/audience. For machine
transcription to work (to be suitable for human view), it needs quite some
requirement to the audio itself (very little noise, speaker accent, etc).
If this is not possible, you'd have to use human, as nick suggested. There are
tons of human transcription companies out there. They typically offer at a
price of $2~3/minute, which is hefty if you are targeting regular people.
Alternatively, you can check out us,
http://nexiwave.com . We offer speech indexing, which
you can let your user to quite accurately locate to the exact location in
their audio, through search. They can then request for a human transcription
if desired. We offer a Web Service API, so the backend has been taken care of.
You only need to worry about the frontend (and integration to human
transcription).
Ben
Hi guys,
I do not know have you solved your problem or not but I wanna recommend a way of free audio to text converter https://audext.com/ You can choose any audio format you have and convert to text it during a few minutes. Hope, I was useful for you in this advice.
Last edit: Alona Serga 2019-09-26