Transcription
Speech-to-text with the language, timed segments and speakers.
Speech-to-text, with the language detected and timed segments. A job with only transcription and metadata gets
the longer audio-only length limit.
curl -X POST https://api.netraflow.com/v1/jobs \
-H "Content-Type: application/json" \
-H "X-Api-Key: sk_live_your_key_here" \
-d '{
"url": "https://www.youtube.com/watch?v=dQw4w9WgXcQ",
"capabilities": ["transcription"]
}'{
"data": {
"job_id": "job_abcdefghij123456",
"status": "completed",
"results": {
"transcription": {
"text": "We're no strangers to love. You know the rules and so do I...",
"segments": [
{
"start": 0.0,
"end": 4.8,
"text": "We're no strangers to love"
},
{
"start": 4.8,
"end": 8.2,
"text": "You know the rules and so do I"
}
],
"language": "en",
"duration_seconds": 212.0,
"word_count": 423
}
}
}
}This video has its own captions, so there are no speaker numbers: see
Creator-provided captions.
Response fields
| Prop | Type | Description |
|---|---|---|
| text | string | Full transcription text. |
| segments | Segment[] | Timed text segments with optional speaker labels. |
| word_count | integer | Total number of words transcribed. |
| language | string (optional) | Detected language code, such as "en". |
| duration_seconds | number (optional) | Audio duration in seconds. |
| speakers_detected | integer (optional) | Number of distinct speakers identified. |
| no_speech | boolean (optional) | True when the audio contains no detectable speech. |
Only text, segments and word_count are always present.
Videos with no speech
A silent or music-only video completes with no_speech: true, text: "" and no segments.
Segment fields
| Prop | Type | Description |
|---|---|---|
| start | number | Segment start time in seconds. |
| end | number | Segment end time in seconds. |
| text | string | Transcribed text for this segment. |
| speaker | integer (optional) | Speaker index, from 0. Absent on caption transcripts. |
Creator-provided captions
Some YouTube transcripts come from the creator's own captions, which have no speakers_detected and no
speaker. Check for the fields rather than relying on them. No transcript has word-level timings.
See also: transcripts for YouTube, TikTok and Instagram, with real results.