Skip to content
Cloudflare Docs
Docs DirectoryAPIsSDKs
Deepgram logo

nova-3

Automatic Speech Recognition • Deepgram • Hosted

Transcribe audio using Deepgram’s speech-to-text model

Model Info
Terms and Licenselink ↗
BatchYes
PartnerYes
Real-timeYes
Unit Pricing$0.0052 per audio minute, $0.0092 per audio minute (websocket)

Supported languages

Nova-3 on Workers AI supports the following languages for transcription:

LanguageCode(s)
Englishen, en-US, en-AU, en-GB, en-IN, en-NZ
Spanishes, es-419
Frenchfr, fr-CA
Germande, de-CH
Hindihi
Russianru
Portuguesept, pt-BR, pt-PT
Japaneseja
Italianit
Dutchnl

Use multi for automatic multilingual detection across all of the languages listed above.

If no language is specified, the model defaults to en-US. For best accuracy, explicitly set the language code matching your audio.

Usage

export default {
  async fetch(request, env, ctx): Promise<Response> {
    const URL = "https://URL_TO_MP3_FILE/audio.mp3";
    const mp3 = await fetch(URL);




    const resp = await env.AI.run("@cf/deepgram/nova-3", {
      "audio": {
        body: mp3.body,
        contentType: "audio/mpeg"
      },
      "detect_language": true
    }, {
      returnRawResponse: true
    });
    return resp;
  },
} satisfies ExportedHandler<Env>;

Parameters

custom_topic_mode
stringenum: extended, strictSets how the model will interpret strings submitted to the custom_topic param. When strict, the model will only return topics submitted using the custom_topic param. When extended, the model will return its own detected topics in addition to those submitted using the custom_topic param.
custom_topic
stringCustom topics you want the model to detect within your input audio or text if present Submit up to 100
custom_intent_mode
stringenum: extended, strictSets how the model will interpret intents submitted to the custom_intent param. When strict, the model will only return intents submitted using the custom_intent param. When extended, the model will return its own detected intents in addition those submitted using the custom_intents param
custom_intent
stringCustom intents you want the model to detect within your input audio if present
detect_entities
booleanIdentifies and extracts key entities from content in submitted audio
detect_language
booleanIdentifies the dominant language spoken in submitted audio
diarize
booleanRecognize speaker changes. Each word in the transcript will be assigned a speaker number starting at 0
dictation
booleanIdentify and extract key entities from content in submitted audio
encoding
stringenum: linear16, flac, mulaw, amr-nb, amr-wb, opus, speex, g729Specify the expected encoding of your submitted audio
extra
stringArbitrary key-value pairs that are attached to the API response for usage in downstream processing
filler_words
booleanFiller Words can help transcribe interruptions in your audio, like 'uh' and 'um'
keyterm
stringKey term prompting can boost or suppress specialized terminology and brands.
keywords
stringKeywords can boost or suppress specialized terminology and brands.
language
stringThe BCP-47 language tag that hints at the primary spoken language. Depending on the Model and API endpoint you choose only certain languages are available.
measurements
booleanSpoken measurements will be converted to their corresponding abbreviations.
mip_opt_out
booleanOpts out requests from the Deepgram Model Improvement Program. Refer to our Docs for pricing impacts before setting this to true. https://dpgr.am/deepgram-mip.
mode
stringenum: general, medical, financeMode of operation for the model representing broad area of topic that will be talked about in the supplied audio
multichannel
booleanTranscribe each audio channel independently.
numerals
booleanNumerals converts numbers from written format to numerical format.
paragraphs
booleanSplits audio into paragraphs to improve transcript readability.
profanity_filter
booleanProfanity Filter looks for recognized profanity and converts it to the nearest recognized non-profane word or removes it from the transcript completely.
punctuate
booleanAdd punctuation and capitalization to the transcript.
redact
stringRedaction removes sensitive information from your transcripts.
replace
stringSearch for terms or phrases in submitted audio and replaces them.
search
stringSearch for terms or phrases in submitted audio.
sentiment
booleanRecognizes the sentiment throughout a transcript or text.
smart_format
booleanApply formatting to transcript output. When set to true, additional formatting will be applied to transcripts to improve readability.
topics
booleanDetect topics throughout a transcript or text.
utterances
booleanSegments speech into meaningful semantic units.
utt_split
numberSeconds to wait before detecting a pause between words in submitted audio.
channels
numberThe number of channels in the submitted audio
interim_results
booleanSpecifies whether the streaming endpoint should provide ongoing transcription updates as more audio is received. When set to true, the endpoint sends continuous updates, meaning transcription results may evolve over time. Note: Supported only for webosockets.
endpointing
stringIndicates how long model will wait to detect whether a speaker has finished speaking or pauses for a significant period of time. When set to a value, the streaming endpoint immediately finalizes the transcription for the processed time range and returns the transcript with a speech_final parameter set to true. Can also be set to false to disable endpointing
vad_events
booleanIndicates that speech has started. You'll begin receiving Speech Started messages upon speech starting. Note: Supported only for webosockets.
utterance_end_ms
booleanIndicates how long model will wait to send an UtteranceEnd message after a word has been transcribed. Use with interim_results. Note: Supported only for webosockets.

API Schemas

{
  "type": "object",
  "properties": {
    "audio": {
      "type": "object",
      "properties": {
        "body": {
          "type": "object"
        },
        "contentType": {
          "type": "string"
        }
      },
      "required": [
        "body",
        "contentType"
      ]
    },
    "custom_topic_mode": {
      "type": "string",
      "enum": [
        "extended",
        "strict"
      ],
      "description": "Sets how the model will interpret strings submitted to the custom_topic param. When strict, the model will only return topics submitted using the custom_topic param. When extended, the model will return its own detected topics in addition to those submitted using the custom_topic param."
    },
    "custom_topic": {
      "type": "string",
      "description": "Custom topics you want the model to detect within your input audio or text if present Submit up to 100"
    },
    "custom_intent_mode": {
      "type": "string",
      "description": "Sets how the model will interpret intents submitted to the custom_intent param. When strict, the model will only return intents submitted using the custom_intent param. When extended, the model will return its own detected intents in addition those submitted using the custom_intents param",
      "enum": [
        "extended",
        "strict"
      ]
    },
    "custom_intent": {
      "type": "string",
      "description": "Custom intents you want the model to detect within your input audio if present"
    },
    "detect_entities": {
      "type": "boolean",
      "description": "Identifies and extracts key entities from content in submitted audio"
    },
    "detect_language": {
      "type": "boolean",
      "description": "Identifies the dominant language spoken in submitted audio"
    },
    "diarize": {
      "type": "boolean",
      "description": "Recognize speaker changes. Each word in the transcript will be assigned a speaker number starting at 0"
    },
    "dictation": {
      "type": "boolean",
      "description": "Identify and extract key entities from content in submitted audio"
    },
    "encoding": {
      "type": "string",
      "description": "Specify the expected encoding of your submitted audio",
      "enum": [
        "linear16",
        "flac",
        "mulaw",
        "amr-nb",
        "amr-wb",
        "opus",
        "speex",
        "g729"
      ]
    },
    "extra": {
      "type": "string",
      "description": "Arbitrary key-value pairs that are attached to the API response for usage in downstream processing"
    },
    "filler_words": {
      "type": "boolean",
      "description": "Filler Words can help transcribe interruptions in your audio, like 'uh' and 'um'"
    },
    "keyterm": {
      "type": "string",
      "description": "Key term prompting can boost or suppress specialized terminology and brands."
    },
    "keywords": {
      "type": "string",
      "description": "Keywords can boost or suppress specialized terminology and brands."
    },
    "language": {
      "type": "string",
      "description": "The BCP-47 language tag that hints at the primary spoken language. Depending on the Model and API endpoint you choose only certain languages are available."
    },
    "measurements": {
      "type": "boolean",
      "description": "Spoken measurements will be converted to their corresponding abbreviations."
    },
    "mip_opt_out": {
      "type": "boolean",
      "description": "Opts out requests from the Deepgram Model Improvement Program. Refer to our Docs for pricing impacts before setting this to true. https://dpgr.am/deepgram-mip."
    },
    "mode": {
      "type": "string",
      "description": "Mode of operation for the model representing broad area of topic that will be talked about in the supplied audio",
      "enum": [
        "general",
        "medical",
        "finance"
      ]
    },
    "multichannel": {
      "type": "boolean",
      "description": "Transcribe each audio channel independently."
    },
    "numerals": {
      "type": "boolean",
      "description": "Numerals converts numbers from written format to numerical format."
    },
    "paragraphs": {
      "type": "boolean",
      "description": "Splits audio into paragraphs to improve transcript readability."
    },
    "profanity_filter": {
      "type": "boolean",
      "description": "Profanity Filter looks for recognized profanity and converts it to the nearest recognized non-profane word or removes it from the transcript completely."
    },
    "punctuate": {
      "type": "boolean",
      "description": "Add punctuation and capitalization to the transcript."
    },
    "redact": {
      "type": "string",
      "description": "Redaction removes sensitive information from your transcripts."
    },
    "replace": {
      "type": "string",
      "description": "Search for terms or phrases in submitted audio and replaces them."
    },
    "search": {
      "type": "string",
      "description": "Search for terms or phrases in submitted audio."
    },
    "sentiment": {
      "type": "boolean",
      "description": "Recognizes the sentiment throughout a transcript or text."
    },
    "smart_format": {
      "type": "boolean",
      "description": "Apply formatting to transcript output. When set to true, additional formatting will be applied to transcripts to improve readability."
    },
    "topics": {
      "type": "boolean",
      "description": "Detect topics throughout a transcript or text."
    },
    "utterances": {
      "type": "boolean",
      "description": "Segments speech into meaningful semantic units."
    },
    "utt_split": {
      "type": "number",
      "description": "Seconds to wait before detecting a pause between words in submitted audio."
    },
    "channels": {
      "type": "number",
      "description": "The number of channels in the submitted audio"
    },
    "interim_results": {
      "type": "boolean",
      "description": "Specifies whether the streaming endpoint should provide ongoing transcription updates as more audio is received. When set to true, the endpoint sends continuous updates, meaning transcription results may evolve over time. Note: Supported only for webosockets."
    },
    "endpointing": {
      "type": "string",
      "description": "Indicates how long model will wait to detect whether a speaker has finished speaking or pauses for a significant period of time. When set to a value, the streaming endpoint immediately finalizes the transcription for the processed time range and returns the transcript with a speech_final parameter set to true. Can also be set to false to disable endpointing"
    },
    "vad_events": {
      "type": "boolean",
      "description": "Indicates that speech has started. You'll begin receiving Speech Started messages upon speech starting. Note: Supported only for webosockets."
    },
    "utterance_end_ms": {
      "type": "boolean",
      "description": "Indicates how long model will wait to send an UtteranceEnd message after a word has been transcribed. Use with interim_results. Note: Supported only for webosockets."
    }
  },
  "required": [
    "audio"
  ]
}