Optimized for quick response

Deepgram Reviews & Product Details

Small-Business(50 or fewer emp.)
Validated Reviewer
Review source: Invitation from G2 on behalf of the seller
Business partner of the seller or seller's competitor, not included in G2 scores.
What do you like best?

Deepgram is a reliable partner for our company in providing critical transcription/diarisation of customer sales calls. Fast turnaround and always improving accuracy Review collected by and hosted on

What do you dislike?

Like anything - accuracy of transcription is always something we are striving to improve on. It would be great if the accuracy curve sped up Review collected by and hosted on

What problems are you solving with the product? What benefits have you realized?

Call transcription and diarisation as part of our conversation intelligence and sales coaching technology. Review collected by and hosted on

Show More
Show Less

Deepgram Overview

What is Deepgram?

Deepgram is the only true end-to-end Deep Learning ASR offering real-time transcription, built to scale for enterprise. We take the heavy lifting out of noisy, multi-speaker, hard to understand audio transcription, so you can focus on getting the insights you need from your voice data. Use it alone or on top of your existing tech and see results in weeks, not months or longer. It’s time to stop settling for out-of-the-box solutions that don’t meet enterprise needs. When speech recognition that’s “good enough for everyone” isn’t good enough for you, try Deepgram. Highlights: • One hour of speech transcription in <1 minute • 90%+ Accuracy • 30%+ More Cost-Effective Benefits: • Drive better business outcomes by processing 100% of your voice data for analysis • Improve business decisions and surface insights with 90%+ accurate voice data transcriptions • Improve voice data availability with transcriptions at 63X real time speeds • Reduce transcription resources and hardware costs by 30%+ Key Features: • Transcription • Audio Timestamps • Confidences • Deep search by phonetics • REST API integration Options: • Speech models for General Purpose, Phone Call and Meeting audio • Keyword boosting for industry terms, unique product names or non standard audio • Trained speech models for maximum accuracy without keyword boosting • Punctuation and Capitalization • Transcripts broken down by person (speaker diarization) • Redaction • Profanity filtering • Multi-channel • Multi-language • Real-time streaming (<0.5 second latency) or batch (63x speedup over alternatives) • On-premise, Virtual Private Cloud or Cloud Deployment Update Languages: English (US, GB, IN, NZ) Spanish Korean French Portuguese (PT, BR) Russian Turkish Hindi (2021) Mandarin (2021) German (2021) Arabic (2021)

Deepgram Details
Languages Supported
Arabic, German, English, French, Hindi, Korean, Portuguese, Russian, Spanish, Turkish
Product Description

Deepgram builds artificial intelligence to recognize speech, search for moments, and categorize audio and video.

How do you position yourself against your competitors?

Deepgram includes the features, scalability, and flexibility that large businesses and software providers selling into the Enterprise need. Here are the areas where legacy speech and voice recognition solutions fall short:
- Ability to accurately transcribe enterprise specific audio (eg. "AT&T" vs. "18")
- Off the shelf general model yields 70% accuracy, and over 95% with training
- Realtime streaming and batch support
- Ability to distinguish between multiple speakers
- Highly scalable and reliable API
- Flexible deployment - Train your model once and deploy on premise or in the cloud

Seller Details
Company Website
1,081 Twitter followers
LinkedIn® Page
9 employees on LinkedIn®

Overview Provided by:
Show More

Deepgram Videos

Deepgram Screenshots

Answer a few questions to help the Deepgram community
Have you used Deepgram before?

Deepgram Reviews

Write a Review
Filter reviews
Showing 2 Deepgram reviews
Showing 2 reviews
Filter Reviews
Filter Reviews
Sort by
Company Size
User Role
All Industries
Already have Deepgram?
Write a Review
Mid-Market(51-1000 emp.)
Validated Reviewer
Verified Current User
Review source: Invitation from G2 on behalf of the seller
What do you like best?

- Accurate Transcripts

- Diarization

- Speed to Return Transcript

- Able to Pull Various File Types Review collected by and hosted on

What do you dislike?

- Sometimes its difficult to get custom features built out Review collected by and hosted on

What problems are you solving with the product? What benefits have you realized?

Call Center Quality Monitoring at Scale using Speech Analytics. We use Deepgram as our Speech-to-Text Engine in several ways.

One way is to enable Call Criteria to handle large contact center volume and use AI to identify the calls that our predictive engine has analyzed as having a potential violation. Then, we use our human quality control analysts to get a much deeper, more thorough picture of the recording to accurate pinpoint and timestamp what happened and where.

More often than not, there are also a lot of false positives and false negatives surrounding speech analytics. We use humans to accuracy check these so we're eliminating these false negatives/positives altogether. Who would want dirty data? Why settle for data that may have that kind of falsities? How can you apply anything programatic to something that isn't accurate? That's why we use humans to verify if the speech analytics flags were accurate or not. Review collected by and hosted on

Show More
Show Less
Mid-Market(51-1000 emp.)
Validated Reviewer
Review source: Invitation from G2 on behalf of the seller
What do you like best?

Our business in data annotation and language services does not have a standard model for all projects so we're constantly looking for extensions to existing AI models to achieve results. One example was a requirement to do automated time stamping per word in Japanese. Deepgram continually upgraded their models throughout the project to support us. Review collected by and hosted on

What do you dislike?

I honestly have nothing to say here. There's nothing that I dislike. Review collected by and hosted on

Recommendations to others considering the product:

Deepgram's approach to Automated Speech Recognition is unique in the industry in that they train models for each discrete customer which provides more accurate results than a general model.

The team is very easy to work with; near zero latency in their response to questions and offer very attractive pricing. Review collected by and hosted on

What problems are you solving with the product? What benefits have you realized?

We have requirements from customers that their data may not leave our domain so Deepgram has provide us and on prem solution.

For key customers with on-going projects Deepgram provides custom model training that tunes specifically for that customer's audio. Review collected by and hosted on

Show More
Show Less

There are not enough reviews of Deepgram for G2 to provide buying insight. Below are some alternatives with more reviews:

Express Scribe Logo
Express Scribe
Express Scribe is professional audio player software for PC or Mac designed to help transcribe audio recordings.
AssemblyAI - Speech to Text API Logo
AssemblyAI - Speech to Text API
We're a team of engineers and researchers, and we're working to give developers and global companies an alternative to big tech companies when it comes to advanced AI solutions.
Microsoft Bing Speech API Logo
Microsoft Bing Speech API
Microsoft Bing Speech API is a cloud-based API that provides advanced algorithms to process spoken language, it allow developers add speech driven actions to their applications including real-time interaction with the user.
Amazon Transcribe Logo
Amazon Transcribe
Amazon Transcribe is an automatic speech recognition (ASR) service that makes it easy for developers to add speech to text capability to their applications. Using the Amazon Transcribe API, you can analyze audio files stored in Amazon S3 and have the service return a text file of the transcribed speech.
Google Cloud Speech-to-Text Logo
Google Cloud Speech-to-Text
Google Cloud Speech-to-Text is a service that enables developers to convert audio to text by applying neural network models in an easy to use API, it recognizes over 80 languages and variants, to support global user base and can transcribe the text of users dictating to an application's microphone, enable command-and-control through voice, or transcribe audio files, among many other use cases.
Microsoft Speaker Recognition API Logo
Microsoft Speaker Recognition API
Microsoft Speaker Recognition API is a cloud-based APIs that provide the most advanced algorithms for speaker verification and speaker identification that can be divided into two categories: speaker verification and speaker identification.
IBM Watson Speech to Text Logo
IBM Watson Speech to Text
IBM Watson Speech to Text is a tool that can be used anywhere if there is a need to bridge the gap between the spoken word and its written form, it uses machine intelligence to combine information about grammar and language structure with knowledge of the composition of an audio signal to generate an accurate transcription.
HTK Logo
Hidden Markov Model Toolkit (HTK) is a portable toolkit for building and manipulating hidden Markov models that is primarily used for speech recognition research although it has been used for numerous other applications including research into speech synthesis, character recognition and DNA sequencing.
Sayint Logo
Sayint, an AI based conversation analytics solution, helps you uncover valuable insights to improve agent performance, enhance customer satisfaction and drive operational efficiencies. Sayint can analyse both real-time and historical communications (voice, chat, email and social feeds)
Scribbl Logo
Record, transcribe, and search across conference calls.
Show More