Audio and text files are submitted via a HTTP multipart form POST. Each file is included as a multipart file attachment within the POST. The service supports several different formats for the audio file. The text must be a plain utf-8 encoded text file.
Request body
The body of the request must contain text and audio file fields.
E.g.
Content-Type: multipart/form-data; boundary=------------------------2e3fcd51d85aad0a
------------------------2e3fcd51d85aad0a
Content-Disposition: form-data; name="audio"; filename="abc.wav"
Content-Type: audio/wav
audio data
------------------------2e3fcd51d85aad0a
Content-Disposition: form-data; name="text"; filename="abc.txt"
Content-Type: text/plain
untimed text data
------------------------2e3fcd51d85aad0a--
In addition to the required audio and text files, a third param file can be provided which includes control values for the formatting of the primary SRT output. The param file is a json object with the following values:
| Parameter | description |
|---|---|
| chars_per_sec | Character timing used when determining the compliance to other supplied limits. |
| chars_per_line | The maximum number of characters that can appear on a line of text. |
| min_line_time | The minimum length, in milliseconds, of a line of text. |
| max_frame_time | The maximum duration of an SRT frame (milliseconds). |
| min_frame_gap | The minimum amount of time between SRT frames (milliseconds). |
| avg_frame_pop_delay | The average amount of time an SRT frame should remain visible after the text for this frame has been spoken. |
| max_lines_per_frame | Maximum number of lines per SRT frame. |
| ellipse_min_silence | If the time between two SRT frames is above this value, ellipses (…) are added to the end of the last line of the current SRT frame and added to the beginning of the first line of the second frame to indicate a continuation |
Sample param JSON file:
{
"chars_per_sec": 25,
"chars_per_line": 42,
"min_line_time": 700,
"max_frame_time": 7000,
"min_frame_gap": 80,
"avg_frame_pop_delay": 480,
"max_lines_per_frame": 2,
"ellipse_min_silence": 1200
}
User request completion callback
When submitting a request, an optional HTTP callback can be specified which will be invoked when the request completes or fails. The callback is an HTTP POST to the exact URL provided. Only one attempt is made to reach the URL provided. The contents of the HTTP request match the response from the Query audio alignment request status and retrieve results method.
Headers
x-requestid- Contains the request_id received when the request was submittedx-status- One of the following indicating the status of the request:
| Status | Description |
|---|---|
| “Completed” | The request has finished processing and the user callback, if provided has been called. |
| “Failed” | An error occurred processing the request. |
x-error- If the request failed, this would contain an error message.x-userdata- This will contain the user data supplied with the original request.
Body
If the status is “Completed” the body of the request contains the default SRT result from the alignment. If the status is “Failed” the body of the request contains the same error message provided in the “x-error” header.