Speech Intelligence Features

Add intelligence to your transcriptions. Extract insights, identify speakers, capture precise timestamps, and customize your output format. Currently available for batch transcriptions.

API Endpoint: https://tb2.shunyalabs.ai/v1/transcriptions

These features are currently in beta.

Build your API request

Select the features you want to enable

Don’t forget to replace YOUR_API_KEY with your own secret key.
curl -X POST "https://tb2.shunyalabs.ai/v1/transcriptions" \
  -H "x-api-key: YOUR_API_KEY" \
  -F "[email protected]" \
  -F "language_code=auto"

Select features:

Set up advanced features

Speaker Diarization

Automatically identify and get speaker-wise transcripts.

Learn more

Speaker Identification

Register voices and automatically recognize speakers by name.

Learn more

Keyword Normalization

Standardize variations of important terms for consistent analysis.

Learn more

Configure languages

Language Identification

Automatically detect which language is being spoken across 200+ languages.

Learn more

Translation

Convert transcripts into 200+ target languages.

Learn more

Transliteration

Convert speech from one script to another for accessibility.

Learn more

Get insights from audio

Summarization

Generate concise summaries of long conversations automatically.

Learn more

Intent Detection

Understand what speakers are trying to accomplish.

Learn more

Sentiment Analysis

Detect specific emotions like joy, anger, or frustration.

Learn more

Emotion Diarization

Track emotions throughout conversations.

Learn more

Word Timestamps

Get precise timing for every word in your transcript.

Learn more

Automate content moderation

Custom Keyword Hashing

Define sensitive terms and automatically hash them in outputs.

Learn more

Profanity Hashing

Automatically detect and redact profanity in transcripts.

Learn more

Get domain-level accuracy

Medical Keyterm Correction

Automatically correct medical terminology and drug names.

Learn more