GNgnani
Speech-to-Text

Speech to Text (REST)

Convert speech audio to text using Vachana's state-of-the-art Speech-To-Text models.

Features

  • Supports 10 Indian languages with dedicated monolingual models
  • Multilingual support for English and Hindi together
  • Automatic audio format conversion

Supported Languages

LanguageCode
Bengalibn-IN
English (India)en-IN
Gujaratigu-IN
Hindihi-IN
Kannadakn-IN
Malayalamml-IN
Marathimr-IN
Punjabipa-IN
Tamilta-IN
Telugute-IN

Model Selection

  • Single language: Uses the dedicated monolingual model for that language
  • Multiple languages: Uses multilingual model (supports en-IN, hi-IN)

Audio Constraints

  • Maximum duration: 30 seconds (Ideal duration is 30 seconds)
  • Supported formats: WAV, MP3, OGG, FLAC, AAC, M4A
  • Sample rate: Automatically converted to 16000 Hz mono

Response

Returns a JSON object containing:

  • success: Operation status (boolean)
  • timestamp: Request timestamp (format: YYYYMMDD_HHMMSS.mmm)
  • transcript: The transcribed text
post/stt/v3

Request

Request body multipart/form-data

audio_filestring binary required

Audio file to transcribe. Supported formats - WAV, MP3, OGG, FLAC, AAC, M4A. Maximum duration - 60 seconds (Ideal duration is 30 seconds).

language_code'bn-IN' | 'en-IN' | 'gu-IN' | 'hi-IN' | 'kn-IN' | 'ml-IN' | 'mr-IN' | 'pa-IN' | 'ta-IN' | 'te-IN' required

Language code for transcription. Use one of the supported language codes.

Supported values: bn-IN, en-IN, gu-IN, hi-IN, kn-IN, ml-IN, mr-IN, pa-IN, ta-IN, te-IN

format'verbatim' | 'transcribe'

Output format for the transcript.

  • verbatim (default) — Returns the raw spoken-form transcript as recognised by the ASR engine. No post-processing is applied.
  • transcribe — Enables Inverse Text Normalization (ITN). Spoken numeric expressions, currency, dates, times, and phone numbers are automatically converted to their written form (e.g. "five thousand rupees" → "₹5,000").
itn_native_numeralsboolean

When format=transcribe, set to true to render digits in the native script of the target language instead of Western Arabic digits (0–9).

For example, with hi-IN: "पाँच हज़ार रुपये" → "₹५,०००" instead of "₹5,000".

Has no effect when format=verbatim.

Example request

{
  "language_code": "hi-IN",
  "format": "transcribe",
  "itn_native_numerals": true
}

Response

Successful transcription

successboolean

Indicates if the transcription was successful

timestampstring

Request timestamp in format YYYYMMDD_HHMMSS.mmm

transcriptstring

The transcribed text from the audio

Changes

No recorded changes to this operation across all 1 revision of this API.