Using Google Speech API

Cancelled Posted 5 years ago Paid on delivery
Cancelled Paid on delivery

This require an app written in PHP(ideally) or NodeJS using real time speech to text APIs from Google or Microsoft to convert spoken words to text and save a mysql database with second by second timestamp

BACKGROUND: We have a web application based on PHP (mysqli extensions) / MySQL / HTML5 / CSS3 / JQuery / JavaScript / AJAX. This application involves recorded and real time videos being played back (HTML5 video) through the web application. This information is being proved as a background of the technology stack involved in our web application. The developer does not have to do any development on this web application. We will take care of any integrations of the code from this project to the web application.

REQUIREMENTS - PHASE 1 of project: As a video is playing inside an HTML5 <video></video> segment, we need to record any words (English for now) being spoken (microphone input only) on a computer or mobile device, in an audio file (MP3) along with the words being converted to text and made visible in a "textarea" field for subsequent update on a MySQL database. It is very important to track the timestamp of the entire session (every second), so that we keep the video synchronized with the audio stream being recorded and the corresponding text transcription being made. The timestamp will be in seconds for the length of the video. While the "textarea" mentioned above simply displays the speech to text content, in the background however every second for the duration of the video (starting with 0 seconds), the system needs to store either nothing or the text version of the word being spoken.

For example if the word "tester" is spoken and translated to text at the 223rd second into the video, then in a two dimensional array the word "tester" would be updated on the 223 second record in the array. Upon "submit" that contents of the array would be updated in a table in the MySQL database. Any open source speech to text APIs (such as Google or Microsoft) that allows speech to text translation for multiple languages, based on developer's recommendation, may be used in this process.

FUTURE PHASES: For your information, this timestamp tracking will be required in a future milestone of the larger program, but the groundwork for that future application must be ready now.

JavaScript MySQL PHP

Project ID: #17203064

About the project

6 proposals Remote project Active 5 years ago