Pre-recorded Audio
Feature Overview
Complete parameter reference for the batch transcription endpoint. Only model is required.
Core parameters
| Parameter | Type | Default | Description |
|---|
model | string | required | Model name. Use zero-indic. |
file | file | one of | Audio/video file upload. |
url | string | one of | Public audio/video URL. |
language_code | string | auto | ISO 639-1 code (hi, ta, en) or full language name. |
output_script | string | auto | Transliterate output. Values: auto, Devanagari, Tamil, Telugu, Kannada, Bengali, Latin, ITRANS. |
word_timestamps | bool | false | Add per-word start/end times and alignment score to every segment. |
project | string | none | Speaker library namespace. Also tags requests for analytics. |
Speaker intelligence parameters
| Parameter | Type | Default | Description |
|---|
enable_diarization | bool | false | Label each segment SPEAKER_00, SPEAKER_01, etc. |
enable_speaker_identification | bool | false | Map SPEAKER_XX labels to registered names. Requires enable_diarization=true. |
enable_emotion_diarization | bool | false | Add emotion to each segment. Requires enable_diarization=true. |
enable_sentiment_analysis | bool | false | Sentiment analysis using Gemini. Result in nlp_analysis.sentiment. |
enable_intent_detection | bool | false | Intent classification using Gemini. Result in nlp_analysis.intent. |
intent_choices | JSON array | none | Constrain intent to specific labels. |
Transcript enrichment parameters
| Parameter | Type | Default | Description |
|---|
enable_summarization | bool | false | Generate transcript summary using Gemini. Result in nlp_analysis.summary. |
summary_max_length | int | 150 | Approximate maximum word count of the summary. |
output_language | string | none | Translate full transcript using Gemini. Result in nlp_analysis.translation. |
enable_keyterm_normalization | bool | false | Normalise domain-specific terms using Gemini. Result in nlp_analysis.normalized_text. |
keyterm_keywords | JSON array | none | Focus normalisation on specific terms. |
Formatting & redaction parameters
| Parameter | Type | Default | Description |
|---|
enable_profanity_hashing | bool | false | Detect and mask profanity using Gemini. Replaces with **** in text and segments. |
hash_keywords | JSON array | none | Mask specific words/phrases using regex (no LLM). Replaces with ****. |