Reviews from AWS Marketplace
0 AWS reviews
-
5 star0
-
4 star0
-
3 star0
-
2 star0
-
1 star0
External reviews
External reviews are not included in the AWS star rating for the product.
It is overall a solid product.
What do you like best about the product?
The timing and the rate at which we can generate multiple transcripts is very useful.
What do you dislike about the product?
The diarization add on is not that great, it should improve its accuracy.
What problems is the product solving and how is that benefiting you?
It helps in getting ASR transcripts for the audio data that we have and improve our work flow in generating transctipts for our use cases.
- Leave a Comment |
- Mark review as helpful
Already Recommending Deepgram To Peers
What do you like best about the product?
Awesome accuracy, speed & price. Great developer experience, from docs to sdk.
What do you dislike about the product?
Lack of local / self hosted models but I understand that's where you make your money so I don't expect that to change.
What problems is the product solving and how is that benefiting you?
Transcribing audio for building various speech-related datasets, i.e. TTS, STT, command recognition. I love using a local whisper jax model for speed & cost but when deepgram can do diarization, formatting, utterances, etc, it makes it very viable... at the price, enticing, even.
Deepgram is great!
What do you like best about the product?
I like how fast it processes speech especially, and the latency is not that bad. One good thing about it is how fast it is to implement in a new or existing project.
What do you dislike about the product?
It's a bit clunky to use, and the documentation could be a lot clearer. I wish the pricing was also a bit easier on the wallet, maybe for institutions or students.
What problems is the product solving and how is that benefiting you?
Converting speech to text for displaying custom fonts for an art project. It's great!
Deepgram's latest models have far surpassed several other competing products for accuracy and speed.
What do you like best about the product?
Very simple API, fast and extremely accurate results. It's easy to see cost and CPU used per job. Signup credit is very generous.
What do you dislike about the product?
At present, I don't think there is a standard command line api. However it was simple to create a small Python wrapper for job submission.
What problems is the product solving and how is that benefiting you?
We have many hours of dictated audio, but much of it was recorded in ambient conditions of high noise. Deepgram allows me to transcribe this audio, with excellent recognition of technical terms, many proper names, places and other special terms. In general, no noise reduction is needed on the input audio. Also, my files have been in several formats: WAV, FLAC, and MP3, at various sample rates and bit depths; so far, Deepgram can recognize most formats automatically and provides interface options for many others.
Very easy to use API for transcriptions
What do you like best about the product?
Very simple API to transcribe audio in different languages.
What do you dislike about the product?
Compared to other tools, Deepgram seems weak with acronyms. For instance, "JSON" becomes "Jay son"
What problems is the product solving and how is that benefiting you?
Getting transcriptions of voice samples.
Reviewing Deepgram for its speed and accuracy
What do you like best about the product?
The speed with which it sends the transcriptions while maintaining accuracy to a good extent.
What do you dislike about the product?
The time stamps that Deepgram supplies along with the transcript are less accurate for professional use.
What problems is the product solving and how is that benefiting you?
Automatic Speech Recognition with high speed and accuracy.
High speed while maintaining accuracy is just the high combination that we needed and Deepgram has provided that to us.
High speed while maintaining accuracy is just the high combination that we needed and Deepgram has provided that to us.
Best Audio Transcribing software
What do you like best about the product?
Deepgram is easy to use and transcribes audio at a very high accuracy level. The most accurate transcript for my audios.
What do you dislike about the product?
I don't have anything specific I dislike about Deepgram considering they superceded my expectations for an audio transcribing.
What problems is the product solving and how is that benefiting you?
Helps me transcribe my audio fast and accurately. This has saved my time as well as increase my productivity in my work.
Fantastic API, good results
What do you like best about the product?
Deepgram's SDK is very good - I used the typescript library and it included all the correct types, descriptive naming, great examples in JSDoc. I was able to get up an running within a few minutes. The 200 mins free is generous - more than enough to dev on your application and get things to a working state.
What do you dislike about the product?
I do wish that their SDK came with some helper methods for converting to common formats like VTT. These are trivial to do, but I would like them to ship a "blessed" version.
What problems is the product solving and how is that benefiting you?
We use Deepgram for Podcast transcription, which is then fed into other AI models to create a summaries and tags of the podcast.
Very easy to use
What do you like best about the product?
I liked how easy it was to use the API. They include a ready example for you to get started, and I didn't have to do any debugging. The example they provided just worked. The service works well too!
What do you dislike about the product?
I would like it to be a bit cheaper, but it is a good service.
What problems is the product solving and how is that benefiting you?
Their voice recognition software is great. It's easy to use, works well, and fits my use case perfectly (voice recognition on demand). It's solved all my problems in that area with an easy to use API.
The easiest way to get started with ASR
What do you like best about the product?
It's currently the only API that can easily transcribe audio and identify speakers.
What do you dislike about the product?
Currently diarization automatically identifies the number of speakers, but it isn't aways right. Would be nice to be able to override it with user input. Also, my attemps to use the front end failed, so I resorted to using the API.
What problems is the product solving and how is that benefiting you?
Automatically transcribing big audio files with human-level performance
showing 1 - 10