Medienwerk
subtitles
Takes the list of speaking turns as /v1/diarize returns it, with start, end and text per entry, and renders a ready subtitle file in SRT or WebVTT. Lines are wrapped at a configurable width, a turn too long for one cue is split into several with timings interpolated across the split, and speaker labels can be prefixed. No audio file goes in and no recogniser is called: whoever already holds a transcript should not have to pay for recognition a second time just to get an SRT file, which is.
POST https://medien.halowerk.com/v1/subtitlesInputs
| Field | Type | Required | Description |
|---|---|---|---|
| segments | array | Yes | The segment list, exactly as /v1/diarize returns it under "turns". Extra fields such as index, words or duration_seconds are accepted and ignored. |
| format | string | No | Output format. srt is the widest supported, vtt is what a browser reads directly. |
| speaker_labels | boolean | No | Prefix each cue with its speaker, as "Speaker 0: ". Segments without a speaker stay unlabelled. |
| max_line_chars | integer | No | Maximum characters per line. 42 is the usual value for broadcast and streaming subtitles. |
| max_lines_per_cue | integer | No | Maximum lines per cue. A turn needing more is split into several cues with interpolated timings. |
Show full input schema
{
"type": "object",
"required": [
"segments"
],
"additionalProperties": false,
"properties": {
"segments": {
"type": "array",
"minItems": 1,
"maxItems": 5000,
"description": "The segment list, exactly as /v1/diarize returns it under \"turns\". Extra fields such as index, words or duration_seconds are accepted and ignored.",
"items": {
"type": "object",
"required": [
"start",
"end",
"text"
],
"additionalProperties": true,
"properties": {
"start": {
"type": "number",
"minimum": 0,
"description": "Start of the segment in seconds."
},
"end": {
"type": "number",
"minimum": 0,
"description": "End of the segment in seconds."
},
"text": {
"type": "string",
"maxLength": 20000,
"description": "Spoken text of the segment."
},
"speaker": {
"type": [
"integer",
"string",
"null"
],
"description": "Speaker label as returned by /v1/diarize. Used only when speaker_labels is true."
}
}
}
},
"format": {
"type": "string",
"enum": [
"srt",
"vtt"
],
"default": "srt",
"description": "Output format. srt is the widest supported, vtt is what a browser reads directly."
},
"speaker_labels": {
"type": "boolean",
"default": false,
"description": "Prefix each cue with its speaker, as \"Speaker 0: \". Segments without a speaker stay unlabelled."
},
"max_line_chars": {
"type": "integer",
"minimum": 20,
"maximum": 120,
"default": 42,
"description": "Maximum characters per line. 42 is the usual value for broadcast and streaming subtitles."
},
"max_lines_per_cue": {
"type": "integer",
"minimum": 1,
"maximum": 4,
"default": 2,
"description": "Maximum lines per cue. A turn needing more is split into several cues with interpolated timings."
}
}
}Published input example
{
"segments": [
{
"start": 0.5,
"end": 3.2,
"text": "Guten Morgen, schoen dass Sie da sind.",
"speaker": 0
},
{
"start": 3.4,
"end": 6.1,
"text": "Danke fuer die Einladung.",
"speaker": 1
}
],
"format": "srt",
"speaker_labels": true
}Result & documentation
Machine-readable documentation of this hub
Source of this page: the published HALOWERK catalog. Updated: 2026-11-17 UTC.
More endpoints in this hub
- Medienwerk0.03 USDC · variable
ad detect
POSThttps://medien.halowerk.com/v1/ad-detectTakes a podcast episode of up to 120 minutes and returns its advertising segments: paid sponsor reads, self-promotion (newsletter, Patreon, merchandise) and cross-promotion of other shows, each with start and end on sentence boundaries, the
- Medienwerk0.004 USDC · variable
audio concat
POSThttps://medien.halowerk.com/v1/audio-concatTakes 2 to 10 audio or video files (each up to 24 MB, together up to 60 minutes) and joins their audio in the given order. Between parts either a pause of chosen length or a crossfade is placed; optionally every part is brought to a common
- Medienwerk0.01 USDC
audio confidence review
POSThttps://medien.halowerk.com/v1/audio-confidence-reviewReturn the least confident recognized passages with exact listening timestamps. Only audio_url is required. Duration is measured locally and capped at20seconds before Deepgram EU processing; training opt-out sent on every request. Failed wo