LinuxQuestions.org

LinuxQuestions.org (/questions/)
-   Programming (https://www.linuxquestions.org/questions/programming-9/)
-   -   speech-to-text (https://www.linuxquestions.org/questions/programming-9/speech-to-text-4175602076/)

schneidz 03-18-2017 12:22 PM
speech-to-text
 
hi, i am curious about speech-to-text. anyone know how to record some sentences and get a text file with the results. i looked at ibm watsons web page but it seems like one would have to pay for something called bluemix after a 1-month trial.

anyone else experiment with this ?

Popov 03-18-2017 02:45 PM
transcribe audio
 
There are many approaches to use speech recognition. ibm watsons is a one of these cloud solution, along the cloud Google, cloud Bing and so on.
But all these cloud services cost some money. It is not clear from your question what do you want to know: free API to write speech to text program or free service to transcribe audio files.

schneidz 03-18-2017 03:09 PM
thanks. at the moment i am looking for a service to transcribe.

schneidz 03-18-2017 06:19 PM
I guess I am looking for something like a hello-world example ?

Popov 03-19-2017 04:41 AM
imb watson api
 
So, look at Primary source: the https://www.ibm.com/watson/developer.../#introduction

schneidz 03-19-2017 04:25 PM
thanx, this seems fairly straightforward.
Code:
curl -X POST -u "{username}":"{password}"
--header "Content-Type: audio/flac"
--data-binary "@audio-file1.flac"
--data-binary "@audio-file2.flac"
"https://stream.watsonplatform.net/speech-to-text/api/v1/recognize?timestamps=true&word_alternatives_threshold=0.9&keywords=%22colorado%22%2C%22tornado%22%2C%22tornadoes%22&keywords_threshold=0.5&continuous=true"
is there a favored speech-to-text api among programmers ?

Popov 03-20-2017 05:38 PM
favorite Speech API
 
I have investigated three speech API Yandex, GOOGle and IBM (but write program only in Google)
Yandex: Russian language support, mp3 file codac, and work in javascript
IBM: - mp3, work in javascript, but limited language (no Russian for me)
Google: powerful cloud service, but no mp3 support and not javascript, but is available for all languages
Google is in beta state, may be it will be developped later


All times are GMT -5. The time now is 02:47 AM.