Sign in

v1

latestOpenAPI 3.1.02026-07-266827131.9 KB
Audio Stream

Start Audio Stream Transcription

Start real-time audio stream transcription with AssemblyAI

post/audio-stream/start

Request body

audio_urlstring required

Audio stream URL (RTMP or RTSP protocol)

language_codestring

Language code for transcription

punctuateboolean

Add punctuation to the transcript

format_textboolean

Format text in the transcript

language_detectionboolean

Enable automatic language detection

language_confidence_thresholdnumber

Confidence threshold for language detection

audio_start_frominteger

Start transcription from this timestamp (milliseconds)

audio_end_atinteger

End transcription at this timestamp (milliseconds)

multichannelboolean

Enable multi-channel audio processing

speech_modelsstring[]

Array of speech models to use

speech_thresholdnumber

Speech detection threshold (0.0-1.0)

disfluenciesboolean

Include disfluencies in the transcript

speaker_labelsboolean

Enable speaker diarization

speakers_expectedinteger

Expected number of speakers

sentiment_analysisboolean

Enable sentiment analysis

entity_detectionboolean

Enable entity detection

auto_highlightsboolean

Enable automatic highlights extraction

content_safetyboolean

Enable content safety detection

iab_categoriesboolean

Enable IAB category classification

auto_chaptersboolean

Enable automatic chapter generation

summarizationboolean

Enable automatic summarization

summary_modelstring

Model to use for summarization

summary_typestring

Type of summary to generate

custom_topicsboolean

Enable custom topic detection

topicsstring[]

Array of custom topics to detect

redact_piiboolean

Redact personally identifiable information

redact_pii_substring

PII redaction substitution method

redact_pii_policiesstring[]

Array of PII policies to apply

redact_pii_audioboolean

Redact PII from audio

redact_pii_audio_qualitystring

Quality setting for PII audio redaction

filter_profanityboolean

Filter profanity from transcript

custom_spellingobject[]

Array of custom spelling corrections

speech_understandingobject

Speech understanding configuration

Example request

{
  "audio_url": "rtmp://example.com/live/audio",
  "language_code": "en"
}

Response

Audio stream started successfully

codestring
messagestring

Example response

{
  "message": "success",
  "data": {
    "task_id": "660e8400-e29b-41d4-a716-446655440001",
    "message": "Audio stream transcription started successfully"
  }
}