TOS SenseFlow offers a variety of preset operators. This topic describes the currently available preset operators.

Preset operator list

TOS‑developed operators

Operator name	Description
Audio transcoding	Converts WAV audio files to MP3 or AAC format and allows you to specify the audio specifications.
Video transcoding	Converts MP4 video files to MP4 or HLS format and allows you to specify the encoding format and video specifications.
Video generation animation	Generates GIF or WebP files from MP4 video files.
Automatic speech recognition	Recognizes speech content in Chinese or English from WAV or MP4 files and delivers results in JSON format.

Operators from other services

Operator name	Description
Subtitle erasure	Removes subtitles from MP4 files and outputs video files without subtitles. tip This operator is provided by AI MediaKit. Make sure you have activated AI MediaKit and granted TOS the necessary permissions. Usage constraints and additional billing apply (charged by AI MediaKit). For more information, see Subtitle removal.

Audio transcoding operator

The audio transcoding operator uses preset templates to convert WAV files to the specified format. The default start time for transcoding is 00:00:00, and the duration matches that of the source file. These settings cannot be modified.

The table below lists the available templates and their details.

Template name	Output format	Bitrate	Bitrate policy	Sample format	Sample rate	Number of channels
mp3-320k	MP3	320 Kbps	Use the source audio bitrate when the source audio bitrate is less than the audio bitrate	-	44100 Hz	2
mp3-192k	MP3	192 Kbps	Use the source audio bitrate when the source audio bitrate is less than the audio bitrate	-	44100 Hz	2
mp3-128k	MP3	128 Kbps	Use the source audio bitrate when the source audio bitrate is less than the audio bitrate	-	44100 Hz	2
mp3-96k	MP3	96 Kbps	Use the source audio bitrate when the source audio bitrate is less than the audio bitrate	-	44100 Hz	2
aac-256k	AAC	256 Kbps	Use the source audio bitrate when the source audio bitrate is less than the audio bitrate	-	44100 Hz	2
aac-160k	AAC	160 Kbps	Use the source audio bitrate when the source audio bitrate is less than the audio bitrate	-	44100 Hz	2
aac-96k	AAC	96Kbps	Use the source audio bitrate when the source audio bitrate is less than the audio bitrate	-	44100 Hz	2
aac-64k	AAC	64 Kbps	Use the source audio bitrate when the source audio bitrate is less than the audio bitrate	-	44100 Hz	2

Video transcoding operator

The video transcoding operator uses preset templates to convert MP4 files to the specified format. The default start time for transcoding is 00:00:00, and the duration matches that of the source file. These settings cannot be modified.

The table below lists the available templates and their details.

MP4 templates

Template name	Output format	Video parameters	Audio parameters
h264-mp4-ld h265-mp4-ld	MP4	Delete video stream: No Encoding format: H.264 or H.265 Bitrate control factor: 0 Bitrate: h264-mp4-ld: 500 kbps h265-mp4-ld: 400 kbps Resolution: 640 px × proportional height Frame rate: -	Delete audio stream: No Encoding format: AAC Bitrate: 80 kbps Sample format: - Sample rate: 44100 Hz Number of channels: -
h264-mp4-sd h265-mp4-sd	MP4	Delete video stream: No Encoding format: H.264 or H.265 Bitrate control factor: 0 Bitrate: h264-mp4-sd: 800 kbps h265-mp4-sd: 500 kbps Resolution: 720 px × proportional height Frame rate: -	Delete audio stream: No Encoding format: AAC Bitrate: 80 kbps Sample format: - Sample rate: 44100 Hz Number of channels: -
h264-mp4-hd h265-mp4-hd	MP4	Delete video stream: No Encoding format: H.264 or H.265 Bitrate control factor: 0 Bitrate: h264-mp4-hd: 2000 kbps h265-mp4-hd: 1200 kbps Resolution: 1280 px × proportional height Frame rate: -	Delete audio stream: No Encoding format: AAC Bitrate: 128 kbps Sample format: - Sample rate: 44100 Hz Number of channels: -
h264-mp4-fhd h265-mp4-fhd	MP4	Delete video stream: No Encoding format: H.264 or H.265 Bitrate control factor: 0 Bitrate: h264-mp4-fhd: 3500 kbps h265-mp4-fhd: 2200 kbps Resolution: 1920 px × proportional height Frame rate: -	Delete audio stream: No Encoding format: AAC Bitrate: 160 kbps Sample format: - Sample rate: 44100 Hz Number of channels: -
h264-mp4-2k h265-mp4-2k	MP4	Delete video stream: No Encoding format: H.264 or H.265 Bitrate control factor: 0 Bitrate: h264-mp4-2k: 4800 kbps h265-mp4-2k: 3000 kbps Resolution: 2048 px × proportional height Frame rate: -	Delete audio stream: No Encoding format: AAC Bitrate: 160 kbps Sample format: - Sample rate: 44100 Hz Number of channels: -
h264-mp4-4k h265-mp4-4k	MP4	Delete video stream: No Encoding format: H.264 or H.265 Bitrate control factor: 0 Bitrate: h264-mp4-4k: 5600 kbps h265-mp4-4k: 4000 kbps Resolution: 3840 px × proportional height Frame rate: -	Delete audio stream: No Encoding format: AAC Bitrate: 160 kbps Sample format: - Sample rate: 44100 Hz Number of channels: -

HLS templates

Template name	Output format	Shard duration	Video parameters	Audio parameters
h264-hls-ld h265-hls-ld	HLS	5000 ms	Delete video stream: No Encoding format: H.264 or H.265 Bitrate control factor: 0 Bitrate: h264-hls-ld: 500 kbps h265-hls-ld: 400 kbps Resolution: 640 px × proportional height Frame rate: -	Delete audio stream: No Encoding format: AAC Bitrate: 80 kbps Sample format: - Sample rate: 44100 Hz Number of channels: -
h264-hls-sd h265-hls-sd	HLS	5000 ms	Delete video stream: No Encoding format: H.264 or H.265 Bitrate control factor: 0 Bitrate: h264-hls-sd: 800 kbps h265-hls-sd: 500 kbps Resolution: 720 px × proportional height Frame rate: -	Delete audio stream: No Encoding format: AAC Bitrate: 80 kbps Sample format: - Sample rate: 44100 Hz Number of channels: -
h264-hls-hd h265-hls-hd	HLS	5000 ms	Delete video stream: No Encoding format: H.264 or H.265 Bitrate control factor: 0 Bitrate: h264-hls-hd: 2000 kbps h265-hls-hd: 1200 kbps Resolution: 1280 px × proportional height Frame rate: -	Delete audio stream: No Encoding format: AAC Bitrate: 128 kbps Sample format: - Sample rate: 44100 Hz Number of channels: -
h264-hls-fhd h265-hls-fhd	HLS	5000 ms	Delete video stream: No Encoding format: H.264 or H.265 Bitrate control factor: 0 Bitrate: h264-hls-fhd: 3500 kbps h265-hls-fhd: 2200 kbps Resolution: 1920 px × proportional height Frame rate: -	Delete audio stream: No Encoding format: AAC Bitrate: 160 kbps Sample format: - Sample rate: 44100 Hz Number of channels: -
h264-hls-2k h265-hls-2k	HLS	5000 ms	Delete video stream: No Encoding format: H.264 or H.265 Bitrate control factor: 0 Bitrate: h264-hls-2k: 4800 kbps h265-hls-2k: 3000 kbps Resolution: 2048 px × proportional height Frame rate: -	Delete audio stream: No Encoding format: AAC Bitrate: 160 kbps Sample format: - Sample rate: 44100 Hz Number of channels: -
h264-hls-4k h265-hls-4k	HLS	5000 ms	Delete video stream: No Encoding format: H.264 or H.265 Bitrate control factor: 0 Bitrate: h264-hls-4k: 5600 kbps h265-hls-4k: 4000 kbps Resolution: 3840 px × proportional height Frame rate: -	Delete audio stream: No Encoding format: AAC Bitrate: 160 kbps Sample format: - Sample rate: 44100 Hz Number of channels: -

Video generation animation operator

The video generation animation operator uses preset templates to convert a specified range of frames from an MP4 file into an animated image.

The following conversion parameters can be used to create GIF or WebP files.

Parameter	Description
Start time	Sets the start time for video conversion.
Intercept duration	Sets the duration of the video segment to be converted.
Animation format	Sets the format of the animated image file. Supported formats: GIF and WebP.
Configuration output width and height	Sets the output width and height. If left unconfigured, the output dimensions will match those of the original video.
Frame rate	Sets the frame rate and scaling mode. Supported scaling modes: crop, stretch, fill, and fit.
Frame extraction interval	Sets the frame sampling interval and the number of sampled frames.

Automatic speech recognition operator

The automatic speech recognition operator uses preset templates to recognize speech content from WAV or MP4 files and delivers the results in JSON format.

The following conversion parameters can be used for speech recognition.

Parameter	Description
Recognize language	Specifies the language to be recognized: auto: Automatic detection. zh: Chinese. en: English.
Single-segment result	Sets whether to deliver the recognition result as a single combined text segment: true: Single segment. false: Multiple segments.
Output format	Sets the output format for the recognition results. Currently, only the JSON format is supported.

Subtitle erasure operator

The subtitle erasure operator uses preset templates to remove subtitles from MP4 files.

The following conversion parameters can be used to remove subtitles.

Parameter	Description
Erase mode	Sets the subtitle removal mode: Subtitle: Removes text detected as subtitles. Text: Removes all detected text, not just subtitles.
Output coding mode	Sets the output video encoding format: Quality Size
Erase region	Sets a region for subtitle removal.

Torch Object Storage