Deepgram
Deepgram is a real-time speech API built for voice agents and call analytics — Nova-3 ships low-latency streaming plus pre-recorded transcription, with diarization, PII redaction, and summarization wired into a single REST + WebSocket surface.
Drop your audio. Transcript in seconds. First transcript free, then $2 a file or $8 = 1,000 min
Nova-3 is Deepgram's current flagship model, available in nova-3-general (multilingual) and nova-3-medical. The platform exposes one REST endpoint for pre-recorded audio (/v1/listen) and a WebSocket for live streaming, with diarization, word-level timestamps, PII redaction, and summarization toggled by query parameters.
Best for real-time voice agents, call-center analytics, live captioning, and meeting tools where p50 streaming latency is the product. New accounts get $200 in free credit with no card; pay-as-you-go pricing runs from $0.0048/min streaming and $0.0077/min pre-recorded on Nova-3 monolingual, with a separate per-minute rate for the Voice Agent API.
What it is
Deepgram's Nova-2 is one of the strongest streaming ASR models on the market, with very low latency and good accuracy on conversational audio. HIPAA-eligible, per-minute pricing competitive with self-hosted for modest volume. Last price check: 2026-04-20.
Watch out for: Lower language coverage than Whisper variants; proprietary.
Install / use
Where Deepgram fits · 6 use-cases
Deepgram's strengths cluster around streaming latency, conversational accuracy, and an API surface that bundles agents + ASR + redaction. Pick the card closest to your build — each links to the canonical docs section.
The Agent API wraps Nova-3 STT, an LLM step, and Deepgram TTS behind one WebSocket so you ship a conversational agent without stitching three providers. Drop-in for phone bots, IVR replacements, and in-product voice copilots.
Built-in turn-taking + barge-in
Batch call recordings through /v1/listen with diarize=true and redact=pii to get speaker-labeled, PII-scrubbed transcripts ready for QA scoring and topic mining. HIPAA-eligible on paid plans.
diarize + redact + summarize=v2
Open a WebSocket to wss://api.deepgram.com/v1/listen and stream PCM; interim transcripts arrive word-by-word with timestamps. Common stack for webinar captions, live-event accessibility, and broadcast workflows.
Word-level timestamps inline
Send the episode URL or upload bytes; ask for diarize=true, punctuate=true, paragraphs=true, and summarize=v2 in one call to get a publishable transcript plus a model-generated recap.
Single request returns all artifacts
nova-3-general handles 10 base languages plus regional variants under one model id — useful when you can't predict the input language, or when you need code-switching inside a single utterance.
Detect + transcribe in one pass
Set redact=pii (or fine-grained tags like numbers, ssn) and the transcript ships with sensitive spans replaced by typed placeholders like [PHONE_NUMBER_1] — raw audio is not retained when the no-store option is enabled on enterprise plans.
Healthcare model for clinical audio
Quickstart · pick a language
Three working ways to transcribe a remote audio URL with Nova-3. Export your key as DEEPGRAM_API_KEY first — grab one from the Deepgram console (free $200 credit, no card).
Official deepgram-sdk v7+ · transcribe any HTTPS audio URL with Nova-3.
# pip install deepgram-sdk
import os
from deepgram import DeepgramClient, PrerecordedOptions
dg = DeepgramClient(os.environ["DEEPGRAM_API_KEY"])
source = {"url": "https://dpgr.am/spacewalk.wav"}
options = PrerecordedOptions(
model="nova-3",
smart_format=True,
diarize=True,
punctuate=True,
summarize="v2",
)
resp = dg.listen.rest.v("1").transcribe_url(source, options)
print(resp.results.channels[0].alternatives[0].transcript)Official @deepgram/sdk · same pre-recorded call from Node 18+.
// npm install @deepgram/sdk
import { createClient } from "@deepgram/sdk";
const dg = createClient(process.env.DEEPGRAM_API_KEY);
const { result, error } = await dg.listen.prerecorded.transcribeUrl(
{ url: "https://dpgr.am/spacewalk.wav" },
{
model: "nova-3",
smart_format: true,
diarize: true,
punctuate: true,
summarize: "v2",
}
);
if (error) throw error;
console.log(result.results.channels[0].alternatives[0].transcript);Plain HTTPS POST to /v1/listen · useful for shell pipelines and edge runtimes.
# pre-recorded URL
curl --request POST \
--url 'https://api.deepgram.com/v1/listen?model=nova-3&smart_format=true&diarize=true&punctuate=true&summarize=v2' \
--header "Authorization: Token $DEEPGRAM_API_KEY" \
--header 'Content-Type: application/json' \
--data '{"url":"https://dpgr.am/spacewalk.wav"}'
# or a local file
curl --request POST \
--url 'https://api.deepgram.com/v1/listen?model=nova-3&smart_format=true' \
--header "Authorization: Token $DEEPGRAM_API_KEY" \
--header 'Content-Type: audio/wav' \
--data-binary @call.wavWhat people actually do with Deepgram-style transcription
The tool is the means. These are the jobs — each one priced at published rates, each one wired up on its own page.
Features
| Speaker diarization | Yes |
| Word-level timestamps | Yes |
| Streaming / real-time | Yes |
| Languages supported | 36 |
| HIPAA eligible | Yes |
Links
- developers.deepgram.com/docs ↗Documentation root — quickstarts, feature guides, and API surface for /v1/listen + agent + TTS.
- Models & languages overview ↗Nova-3 variants (general / medical), supported languages, and which features each model exposes.
- Voice Agent getting-started ↗WebSocket Agent API — ASR + LLM + TTS in one stream, with code in Python, JS, C#, Go.
- Live streaming audio guide ↗Real-time transcription over WebSocket with interim partial results and word timestamps.
- deepgram/deepgram-python-sdk ↗Official Python SDK — v7.x line, async + sync clients, pre-recorded + streaming + agent.
- deepgram/deepgram-js-sdk ↗Official JavaScript / TypeScript SDK — browser, Node 18+, edge runtimes.
- deepgram/deepgram-go-sdk ↗Official Go SDK — same feature surface as Python / JS.
- deepgram.com/pricing ↗Current per-minute rates for Nova-3 streaming / pre-recorded plus the Voice Agent SKU and Growth plan discounts.
- status.deepgram.com ↗Live status for the Public, Batch, Streaming, TTS, and Voice Agent APIs — subscribe via email / Slack / webhook.
- console.deepgram.com/signup ↗Free $200 credit, no card required — keys appear in the console immediately after signup.
Deepgram vs Whipscribe
| Feature | Deepgram | Whipscribe |
|---|---|---|
| Category | Transcription APIs | Transcription APIs |
| Pricing | Not verified | $8–$24 one-time packs (credits never expire) · $2 single unlock · free instant preview |
| Speaker diarization | Not verified | Yes |
| Word timestamps | Not verified | Yes |
| Streaming | Not verified | No |
| Languages | 36 | 99 |
| Platforms | API | Web, API, MCP |
Where this category is heading
From the vendor changelogs we track weekly — what changed in August 2026, and what it means if you are choosing now.
AssemblyAI moved summarisation onto an LLM this month; every vendor is racing to return action items, quotes and topics with the text rather than as an add-on.
Whipscribe today Every Whipscribe job already returns an insights payload — summary, key quotes, topics and speakers — from the same job id, at no extra charge.
Deepgram shipped self-hosted container images in August — the market is moving toward audio that stays inside a boundary the customer controls, because teams with customer calls or unreleased material are refusing shared model endpoints.
Whipscribe today Whipscribe runs on our own GPUs in a private, secured cloud. Audio is never forwarded to OpenAI or any third-party model.
The fastest-growing way to use a transcription API is not a form — it is Claude, Cursor or a workflow runner calling it mid-task through MCP.
Whipscribe today Whipscribe ships an MCP server: transcribe, search and summarise from an assistant without wiring anything.
AssemblyAI's 1.0 SDK unified async, realtime and sync; Deepgram's CLI went to 0.3. The unit of work is becoming the folder or the bucket, not the file.
Whipscribe today Submit with an Idempotency-Key and a batch_id, poll by job, retry safely. The S3 connector runs a whole prefix in one grant.
Deepgram added Afrikaans, Georgian and Armenian and improved a dozen more this month. Coverage is widening while quality still clusters around English and the large European languages.
Whipscribe today 99+ languages auto-detected. Ask for a language explicitly when you know it — auto-detect on a short or noisy clip is the most common cause of a wrong-language transcript.
Source: Deepgram and AssemblyAI changelogs, scanned 2026-08-24.
Alternatives to Deepgram
Frequently asked about Deepgram
What is Deepgram?
Nova-3 is Deepgram's current flagship model, available in nova-3-general (multilingual) and nova-3-medical. The platform exposes one REST endpoint for pre-recorded audio (/v1/listen) and a WebSocket for live streaming, with diarization, word-level timestamps, PII redaction, and summarization toggled by query parameters.
How much does Deepgram cost?
Deepgram is a paid product — published pricing: from $0.0043/min. Pricing changes; verify on the vendor's own page before budgeting.
What platforms does Deepgram support?
Deepgram is an API — you call it from whatever you build, on any platform with an HTTP client.
How do I get started with Deepgram?
Deepgram is used through its API: get a key from the vendor, then call it from your own code. There is no desktop app to install.
How many languages does Deepgram support?
Deepgram lists 36 languages.
What are the limitations of Deepgram?
Lower language coverage than Whisper variants; proprietary.
Who is Deepgram best for?
Real-time voice apps (agents, meeting tools) where streaming latency is the product.
What should I know before choosing Deepgram?
Best for real-time voice agents, call-center analytics, live captioning, and meeting tools where p50 streaming latency is the product. New accounts get $200 in free credit with no card; pay-as-you-go pricing runs from $0.0048/min streaming and $0.0077/min pre-recorded on Nova-3 monolingual, with a separate per-minute rate for the Voice Agent API.
Is Deepgram open source?
No. Deepgram is a proprietary API. If you need source you can read and run yourself, the open-source tools in this directory are the place to look.
What kind of tool is Deepgram?
In this directory Deepgram is filed under commercial api as a API.
What are the alternatives to Deepgram?
There is a side-by-side page at /tools/deepgram-alternatives comparing Deepgram with the closest tools in the same category on price, platform and features.
Are there setup recipes for Deepgram?
Yes — this page carries 3 tested setups: Python SDK · pre-recorded URL; Node / JavaScript SDK; cURL · no SDK.
Can I automate Deepgram-style transcription without running it myself?
Yes. If what you want is transcripts rather than the tool itself, Whipscribe does the same job as a hosted API: submit a file or URL, poll a job id, pull the result as txt, json, srt, vtt or docx. $0.008 a minute — about $0.48 an audio hour — bought as credits that never expire. The automation recipes on this site show it wired to Drive folders, Zoom recordings, S3 buckets and no-code platforms.
How is Whipscribe different from Deepgram?
Deepgram is another vendor's product; Whipscribe is a per-minute transcription service. We run the models on our own GPUs in a private cloud — audio is never forwarded to OpenAI or any third-party model — and every job returns five formats plus an AI summary, quotes and topics. Your first transcript in the web app is free at any length, no card.
What does transcription cost if I use Whipscribe instead of Deepgram?
$0.008 a minute — about $0.48 an audio hour — bought as credits that never expire. There is no subscription and no monthly minimum: a 3-hour recording is about $1.44, a thousand-hour archive is about $480 once. Your first transcript in the web app is free at any length, no card, so you can check accuracy on your own audio before paying anything.
Whipscribe is a managed faster-whisper + whisperX service. If you want transcripts without running infrastructure, paste a URL or drop a file in the form below — you'll have a transcript in seconds.
Explore
All transcription tools · Audio technology hub · Transcribe any platform · Audio & video formats · How-to guides · Glossary · Playbooks · Apps · Broadcast & radio · Podcast transcripts · Use cases · Blog · Transcription API · Integrations · Automations · For your industry