Try Live STT Models

Compare all STT

Test Deepgram Nova-3 Live

Upload your own audio file (not huge for now) and get an instant transcript from Deepgram Nova-3. No login required.

Try Deepgram Nova-3 on your audio

Drop a file below. We have pre-selected Deepgram Nova-3 for you.

Input Source

Click or drag audio file here

Supports MP3, M4A, WAV, OGG

Max file size: 100MB

Configuration

Advanced Options

Enables raw parameter access for Deepgram Nova-3. Disables universal options.

Not supported by all selected

Provides a hint for the minimum and maximum number of expected speakers to improve diarization accuracy.

Boosts the recognition probability of specific words or phrases, such as proper nouns or domain-specific terms. Provide one phrase per line.

Note: Files and transcripts are not stored on our servers and are used only to complete your request. More features are coming.

Technical Specifications

Configurable Parameters

These universal options are mapped to provider-specific features.

languagestring

Language

Primary language of the audio

Capabilities

  • Diarization
  • Profanity_filter
  • Punctuation
  • Smart_formatting
  • Word_boost

Native Configuration

These are the provider's native API parameters — shown exactly as exposed by the vendor.

diarize

Recognize speaker changes. Each word will be assigned a speaker number starting at 0.

punctuate

Add punctuation and capitalization to the transcript.

paragraphs

Splits audio into paragraphs to improve transcript readability.

utterances

Segments speech into meaningful semantic units with speaker attribution.

smart_format

Apply formatting to improve readability (dates, times, currency, phone numbers).

filler_words

Transcribe interruptions in your audio like 'uh' and 'um'.

numerals

Convert numbers from written format to numerical format (e.g., 'twenty' to '20').

measurements

Convert spoken measurements to their corresponding abbreviations (e.g., 'meters' to 'm').

dictation

Format transcript with dictated speech (e.g., 'comma', 'period', 'new paragraph').

profanity_filter

Convert profanity to the nearest non-profane word or remove it completely.

redact

Redact sensitive information from transcripts. Select common entity groups.

replace

Search for terms/phrases and replace them. Format: 'find:replace' (e.g., 'Acme:Company').

search

Search for specific terms or phrases in the audio. Results include timestamps and confidence scores.

keyterms

Boost or suppress specialized terminology and brands (Nova-3 only). Format: 'term' to boost, '-term' to suppress.

detect_entities

Identify and extract key entities like names, dates, locations, organizations.

sentiment

Analyze sentiment throughout the transcript (positive, negative, neutral).

summarize

Generate a summary of the audio content.

topics

Detect topics throughout the transcript.

intents

Recognize speaker intent throughout the transcript.

multichannel

Transcribe each audio channel independently.

detect_language

Automatically detect the dominant language spoken in the audio.

utt_splitDefault: 0.8

Seconds to wait before detecting a pause between words (default: 0.8).

tag

Label your requests for identification during usage reporting.

About Deepgram Nova-3

Deepgram Nova-3 model for fast, accurate speech-to-text with diarization and audio intelligence options

Pricing

A detailed pricing breakdown will be available here shortly. For now, please refer to the provider's official website.

View Deepgram Nova-3 official documentation

🚀