Deepgram Speech Understanding

Accurately identify, extract, and summarize conversational audio to deliver amazing customer experiences.

It’s Natural Language Understanding (NLU) built on the industry’s most accurate, reliable speech-to-text.

Explore Docs

UNDERSTANDING

Understand more, right out of the box.

Move beyond just transcription, without hiring additional experts.

  • Speaker diarization

    Know who’s talking. Detect and label speaker changes throughout a conversation with speaker diarization.

  • Entity detection

    Identify a variety of entities spoken in audio such as names, locations, account numbers, and more with entity detection.

  • Summarization

    Summarize sections of content in your audio for better readability, ease of use, and analysis with Summarization.

  • Topic detection

    Use topic detection to identify and label key topics from your audio for ease of search and to discover trends and insights.

  • Language translation

    Convert your transcript into dozens of available languages such as Spanish, French, Hindi, and many others with language translation.

  • Language detection

    Use language detection to identify the dominant language in your audio and transcribe in the identified language.

  • Sentiment analysis

    Determine positive, neutral, or negative opinions in the transcript of your audio for better CX, brand health, and more with sentiment analysis.

Explore more features

Try it Free

00:00/00:00

diarize=true

Speaker 0: Alright. I’m ready.

Speaker 1: Good evening. I’m Dr. Emmett Brown. I’m standing on the parking lot at Twin Pines Mall. It’s Saturday Morning October twenty sixth nineteen eighty five one eighteen AM. And this is temporal experiment number one. Come on Einey. Hey, boy. Get in there. At a boy. In you go. Sit down. Get your seatbelt on. That’s it.

Speaker 0: Okay.

Speaker 1: Please note, that Einstein’s clock is in precise synchronization with my control watch. Got it?

Speaker 0: Right. Check doc.

Speaker 1: Good. Have a good trip einstein. Watch your head.

Speaker 0: You got that thing hooked up to the car?

Speaker 1: Watch this.

Speaker 0: Yeah Ok.

Speaker 1: Not me the car, the car. If my calculations are correct. When this baby hits eighty eight miles per hour, you’re gonna see some serious s**t. Watch this watch this. What did I tell you? Eighty eight miles per hour. The thermal displacement occurred exactly what? One O two A M and zero seconds.

Speaker 0: Jesus Christ. Jesus Christ, doc, you disintegrated einstein.

Speaker 1: Calm down Marty. I didn’t disintegrate anything. The molecular structure of both Einstein and the car are completely intact.

Speaker 0: Then where hell are they?

Speaker 1: The appropriate question is, when the hell are they? You see, Einstein has just become the world’s first time traveler. I set him into the future. One minute into the future to be exact. Now precisely one twenty one AM and zero seconds we shall catch up with him and the time machine.

Speaker 0: Wait a minute. Wait a minute. Doc. Are you telling me that you built a time machine out of a Delorean?

detect_entities=true

Alright. I’m ready. Good evening. I’m Dr. Emmett Brown. I’m standing on the parking lot at Twin Pines Mall. It’s Saturday Morning October 26th, 1985 1:18 AM. And this is temporal experiment number 1. Come on Einey. Hey, boy. Get in there. At a boy. In you go. Sit down. Get your seatbelt on. That’s it. Okay. Please note, that Einstein’s clock is in precise synchronization with my control watch. Got it? Right. Check Doc. Good. Have a good trip Einstein. Watch your head. You got that thing hooked up to the car? Watch this. Yeah Ok. Not me the car, the car. If my calculations are correct. When this baby hits 88 miles per hour, you’re gonna see some serious s**t. Watch this watch this. What did I tell you? 88 miles per hour. The thermal displacement occurred exactly what? 1:02 AM and zero seconds. Jesus Christ. Jesus Christ, Doc, you disintegrated Einstein. Calm down Marty. I didn’t disintegrate anything. The molecular structure of both Einstein and the car are completely intact. Then where hell are they? The appropriate question is, when the hell are they? You see, Einstein has just become the world’s first time traveler. I set him into the future. 1 minute into the future to be exact. Now precisely 1:21 AM and 0 seconds we shall catch up with him and the time machine. Wait a minute. Wait a minute. Doc. Are you telling me that you built a time machine out of a Delorean?

| Person
| Location
| Date
| Time
| Cardinal number
| Quantity
| Organization

summarize=true&punctuate=true

Alright. I’m ready. Good evening. I’m Dr. Emmett Brown. I’m standing on the parking lot at Twin Pines Mall. It’s Saturday Morning October twenty sixth nineteen eighty five one eighteen AM. And this is temporal experiment number one. Come on Einey. Hey, boy. Get in there. At a boy. In you go. Sit down. Get your seatbelt on. That’s it. Okay. Please note, that Einstein’s clock is in precise synchronization with my control watch. Got it? Right. Check Doc. Good. Have a good trip Einstein. Watch your head. You got that thing hooked up to the car? Watch this. Yeah Ok. Not me the car, the car. If my calculations are correct. When this baby hits eighty eight miles per hour, you’re gonna see some serious s**t.

“summary”: “An experiment is being conducted. The speaker is Dr. Emmett Brown and he gives his location and the date and time. Someone is traveling by car and the experiment is about to begin.”

Watch this watch this. What did I tell you? Eighty eight miles per hour. The thermal displacement occurred exactly what? One O two AM and zero seconds. Jesus Christ. Jesus Christ, doc, you disintegrated Einstein. Calm down Marty. I didn’t disintegrate anything. The molecular structure of both Einstein and the car are completely intact. Then where hell are they? The appropriate question is, when the hell are they? You see, Einstein has just become the world’s first time traveler. I set him into the future. One minute into the future to be exact. Now precisely one twenty one AM and zero seconds we shall catch up with him and the time machine. Wait a minute. Wait a minute. Doc. Are you telling me that you built a time machine out of a Delorean?

“summary”: “There is concern over the traveler’s safety but everything is intact. The event is the world’s first time travel experiment made out of a Delorean.”

 

detect_topics=true&punctuate=true

Alright. I’m ready. Good evening. I’m Dr. Emmett Brown. I’m standing on the parking lot at Twin Pines Mall. It’s Saturday Morning October twenty sixth nineteen eighty five one eighteen AM. And this is temporal experiment number one. Come on Einey. Hey, boy. Get in there. At a boy. In you go. Sit down. Get your seatbelt on. That’s it. Okay. Please note, that Einstein’s clock is in precise synchronization with my control watch. Got it? Right. Check Doc. Good. Have a good trip Einstein. Watch your head. You got that thing hooked up to the car? Watch this. Yeah Ok. Not me the car, the car. If my calculations are correct. When this baby hits eighty eight miles per hour, you’re gonna see some serious s**t.

“topics”: “introduction, experiment, driving”

Watch this watch this. What did I tell you? Eighty eight miles per hour. The thermal displacement occurred exactly what? One O two AM and zero seconds. Jesus Christ. Jesus Christ, doc, you disintegrated Einstein. Calm down Marty. I didn’t disintegrate anything. The molecular structure of both Einstein and the car are completely intact. Then where hell are they? The appropriate question is, when the hell are they? You see, Einstein has just become the world’s first time traveler. I set him into the future. One minute into the future to be exact. Now precisely one twenty one AM and zero seconds we shall catch up with him and the time machine. Wait a minute. Wait a minute. Doc. Are you telling me that you built a time machine out of a delorean?

“topics”: “speed, physics, time travel”

translate=true

Alright. I’m ready. Good evening. I’m Dr. Emmett Brown. I’m standing on the parking lot at Twin Pines Mall. It’s Saturday Morning October twenty sixth nineteen eighty five one eighteen AM. And this is temporal experiment number one. Come on Einey. Hey, boy. Get in there. At a boy. In you go. Sit down. Get your seatbelt on. That’s it. Okay. Please note, that Einstein’s clock is in precise synchronization with my control watch. Got it? Right. Check Doc. Good. Have a good trip Einstein. Watch your head. You got that thing hooked up to the car? Watch this. Yeah Ok. Not me the car, the car. If my calculations are correct. When this baby hits eighty eight miles per hour, you’re gonna see some serious s**t. Watch this watch this. What did I tell you? Eighty eight miles per hour. The thermal displacement occurred exactly what? One O two AM and zero seconds. Jesus Christ. Jesus Christ, doc, you disintegrated Einstein. Calm down Marty. I didn’t disintegrate anything. The molecular structure of both

Bien. Estoy listo. Buenas noches. Soy el Dr. Emmett Brown. Estoy parado en el estacionamiento del Twin Pines Mall. Es sábado por la mañana, veintiséis de octubre de mil novecientos ochenta y cinco, uno dieciocho de la mañana. Y este es el experimento temporal número uno. Vamos Einey. Oye, chico. Entra allí. a un chico. En ti vas. Siéntate. Ponte el cinturón de seguridad. Eso es todo. Bueno. Tenga en cuenta que el reloj de Einstein está en sincronización precisa con mi reloj de control. ¿Entendido? Correcto. Listo, Doc. Bien. Que tengas buen viaje Einstein. Cuidado con tu cabeza. ¿Tienes esa cosa conectada al coche? Ve esto. Sí, ok. No yo el coche, el coche. Si mis cálculos son correctos. Cuando este bebé llegue a ochenta y ocho millas por hora, vas a ver una m****a seria. Mira esto mira esto. ¿Qué te dije? Ochenta y ocho millas por hora. El desplazamiento térmico ocurrió exactamente a la una y veintidós AM y cero segundos. Jesucristo. Jesucristo, doc, desintegraste a Einstein. Tranquilo Marti. No desintegré nada. La estructura molecular tanto de Einstein como del automóvil están completamente intactas. Entonces, ¿dónde diablos están? La pregunta apropiada es, ¿cuándo diablos están? Verás, Einstein acaba de convertirse en el primer viajero en el tiempo del mundo. Lo puse en el futuro. Un minuto en el futuro para ser exactos. Ahora, exactamente a la una y veintiuna de la mañana y cero segundos, lo alcanzaremos a él y a la máquina del tiempo. Espera un minuto. Espera un minuto. Doc. ¿Me estás diciendo que construiste una máquina del tiempo con un DeLorean?

tier=enhanced&detect_language=true

“detected_language”: “en”

Alright. I’m ready. Good evening. I’m Dr. Emmett Brown. I’m standing on the parking lot at Twin Pines Mall. It’s Saturday Morning October twenty sixth nineteen eighty five one eighteen AM. And this is temporal experiment number one. Come on Einey. Hey, boy. Get in there. At a boy. In you go. Sit down. Get your seatbelt on. That’s it. Okay. Please note, that Einstein’s clock is in precise synchronization with my control watch. Got it? Right. Check Doc. Good. Have a good trip Einstein. Watch your head. You got that thing hooked up to the car? Watch this. Yeah Ok. Not me the car, the car. If my calculations are correct. When this baby hits eighty eight miles per hour, you’re gonna see some serious s**t. Watch this watch this. What did I tell you? Eighty eight miles per hour. The thermal displacement occurred exactly what? One O two AM and zero seconds. Jesus Christ. Jesus Christ, doc, you disintegrated Einstein. Calm down Marty. I didn’t disintegrate anything. The molecular structure of both Einstein and the car are completely intact. Then where hell are they? The appropriate question is, when the hell are they? You see, Einstein has just become the world’s first time traveler. I set him into the future. One minute into the future to be exact. Now precisely one twenty one AM and zero seconds we shall catch up with him and the time machine. Wait a minute. Wait a minute. Doc. Are you telling me that you built a time machine out of a Delorean?

sentiment_analysis=true

Alright. I’m ready. Good evening. I’m Dr. Emmett Brown. I’m standing on the parking lot at Twin Pines Mall. It’s Saturday Morning October twenty sixth nineteen eighty five one eighteen AM. And this is temporal experiment number one.

“sentiment”: “neutral” 😐

Come on Einey. Hey, boy. Get in there. At a boy. In you go. Sit down. Get your seatbelt on. That’s it. Okay. Please note, that Einstein’s clock is in precise synchronization with my control watch. Got it? Right. Check Doc. Good. Have a good trip Einstein. Watch your head.

“sentiment”: “positive” 😀

You got that thing hooked up to the car? Watch this. Yeah Ok. Not me the car, the car. If my calculations are correct. When this baby hits eighty eight miles per hour, you’re gonna see some serious s**t.

“sentiment”: “positive” 😀

Watch this watch this. What did I tell you? Eighty eight miles per hour. The thermal displacement occurred exactly what? One O two AM and zero seconds.

“sentiment”: “positive” 😀

Speech recognition understanding features

Speaker diarization

Know who’s talking. Detect and label speaker changes throughout a conversation.

View speaker diarization

Summarization

Summarize sections of content in your audio for better readability, ease of use, and analysis.

View summarization

Topic detection

Identify and label key topics for ease of classification and to discover trends and insights.

View topic detection

Language detection

Automatically identify and transcribe the dominant language in your audio.

View language detection

Entity detection

Identify a variety of entities such as names, locations, account numbers, quantities, and more.

Coming soon

Language translation

Convert your transcript into dozens of available languages such as Spanish, French, Hindi, and many others.

Coming soon

Sentiment analysis

Determine positive, neutral, or negative opinions to identify warm leads or problematic calls.

Coming soon

 

Your topic modeling, your sentiment analysis, your detection of product features and competitors, and value drivers all come from the transcript. The quality of your transcript therefore has implications all the way down your value chain.”

 

NLU built on reliable transcription foundation

From high fidelity, single-speaker dictation to staticky, acronym-heavy ground-to-space communications, Deepgram delivers accurate transcriptions for reliable speech understanding even on the toughest, real-world audio.

View Transcription

Build with what you want.

Use our Python, Node.js, .NET SDKs, or our REST API to get transcripts that read like a human wrote them.

Read the Docs

from deepgram import Deepgram
import asyncio
import json

DEEPGRAM_API_KEY = "YOUR_DEEPGRAM_API_KEY"


async def main():

    # Initialize the Deepgram SDK
    deepgram = Deepgram(DEEPGRAM_API_KEY)

    FILE = 'URL_TO_YOUR_FILE'

    source = {
        'url': FILE
    }

    response = await asyncio.create_task(
        deepgram.transcription.prerecorded(
            source
        )
    )

    print(json.dumps(response, indent=4))

asyncio.run(main())
{
  "channel_index": [
    {
      "channel": "string",
      "num_channels": 0
    }
  ],
  "duration": 0,
  "start": "string",
  "is_final": true,
  "channel": {
    "alternatives": [
      {
        "transcript": "string",
        "confidence": 0,
        "words": [
          {
            "word": "string",
            "start": 0,
            "end": 0,
            "confidence": 0
          }
        ]
      }
    ]
  }
}
const { Deepgram } = require('@deepgram/sdk')

const deepgram = new Deepgram(DEEPGRAM_API_KEY)
const audioSource = { url: URL_OF_FILE }

const response = await deepgram.transcription.preRecorded(audioSource)

console.dir(response, {depth: null})
{
  "channel_index": [
    {
      "channel": "string",
      "num_channels": 0
    }
  ],
  "duration": 0,
  "start": "string",
  "is_final": true,
  "channel": {
    "alternatives": [
      {
        "transcript": "string",
        "confidence": 0,
        "words": [
          {
            "word": "string",
            "start": 0,
            "end": 0,
            "confidence": 0
          }
        ]
      }
    ]
  }
}
using Deepgram;

var deepgram = new DeepgramClient(new Credentials("YOUR_DEEPGRAM_API_KEY"));
var response = await deepgram.Transcription.Prerecorded.GetTranscriptionAsync(
    new UrlSource("URL_TO_FILE"));
{
  "channel_index": [
    {
      "channel": "string",
      "num_channels": 0
    }
  ],
  "duration": 0,
  "start": "string",
  "is_final": true,
  "channel": {
    "alternatives": [
      {
        "transcript": "string",
        "confidence": 0,
        "words": [
          {
            "word": "string",
            "start": 0,
            "end": 0,
            "confidence": 0
          }
        ]
      }
    ]
  }
}

Deploy on-prem, cloud, or VPC.

Our standard deployment is within our cloud, but for more sensitive voice and transcription data, we also offer an on-premises installation or a private cloud installation, where you can control the entire environment. Deepgram is Kubernetes-ready with Docker images, and has pre-built VM images to enable rapid deployment to most cloud providers. Train models and deploy anywhere – on premises or in the cloud.

View Security

See what developers are saying about Deepgram.

The Deepgram API covers the languages we need (and then some), integrates easily with our audio source, is accurate enough, and delivers results quickly. The documentation made it easy to design our code, and the very helpful support engineers were quick to respond to questions and to help us debug our initial efforts.

The speed and accuracy of Deepgram API is the best I have seen.

We provide Fraud Detection services to the insurance industry using intelligent and compliant AI-driven Digital Speech DNA solutions over Blockchain. Using Deepgram allowed us to process a large volume of data quickly and accurately. In addition, Deepgram has the ability to detect different accents which improved the overall accuracy of our scoring module.

The low latency of the response with high accuracy from the websocket connection is the most distinguishing feature from other providers. If this feature was not there then it’s yet another Speech to Text service. I really love the community around it and the team which is driving it, kudos to the DevRel team.

Great speech-to-text results in seconds.

As a software developer, there is plenty to like about Deepgram – complete and easy to follow documentation; easy to use API that allows for quick language-independent implementation; great follow-up support; multiple models including one specifically for telephone-based dictation; not only one of the best but also one of the least expensive speech rec services available; a generous free number of credits are provided at sign-up – plenty enough for experimentation and testing of your application.

We have tested a number of transcription APIs, and Deepgram has consistently come out as the most accurate for our use case. whilst offering a nice Python interface for batch operations. The API schemas are also excellent.

The Best Audio Transcription Service in the Wild!

I have been using Deepgram’s API for a couple of months now, and I am beyond impressed with the accuracy. It is so much better than other voice recognition services that I have tried in the past. I love that it supports so many languages, which is perfect for me because I work with clients worldwide. The best part is that its API is pretty intuitive, which means it doesn’t require any training, which saves me tons of time. I would recommend this to anyone who needs a speech-to-text service!

The ease of use! The simple but powerful APIs make it so quick and easy to start creating something. Not only were the tools very easy to use but they were also incredibly fast and accurate. I came across no translation issues when using the product despite testing it in noisy and non-optimal conditions. And the results were almost instantaneous. Other tools I had looked at were either very restrictive or not very accurate so it was refreshing to find an SDK that gave the flexibility to do whatever I want without compromising on speech and accuracy.

An Automated Speech API with Intuitive Documentation

My favorite part about using Deepgram was the ease of learning. The API documentation is complete and intuitive, and the tutorials in the console left me feeling confident that I could use the API and SDK in either Node or Python projects.

The key to better NLP is better speech-to-text.

And the key to better speech-to-text is end-to-end deep learning. See what faster, more accurate, more scalable ASR can do.

View Transcription