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.
Powerful Transcription Technology
What do you like best about the product?
The customer service was incredibly supportive and responsive. Deepgram's pricing is very competitive, and the transcriptions are both accurate and incredibly fast.
What do you dislike about the product?
I really enjoyed using Deepgram, and there weren't any things I deeply disliked about them. Perhaps there weren't as many paramaters for analysis as other speech to text providers.
What problems is the product solving and how is that benefiting you?
We use deepgram's voice recognition to analyze classroom environments and provide detailed and accurate transcriptions with voice analysis. We're impressed with the quality and time-effectivness of Deepgram.
- Leave a Comment |
- Mark review as helpful
Deepgram is easy and inexpensive, but transcription quality is far below a human-edited transcript
What do you like best about the product?
The price is reasonable, the documentation is mostly excellent, and a basic integration is easy to get going. The API is a pleasure to work with, and the backend is visually well-designed.
What do you dislike about the product?
Very minor issues: Usage monitoring could be a little better. I've also found a few spots where the documentation was out-of-date or vague.
More significant issues: The transcription quality is comparable to other services I've used, but still nowhere near good enough to transcribe our podcasts reliably without a LOT of editing. The "keywords" feature can help a little, but it's not perfect. I also had to jump through many hoops to use entire phrases as keywords (such as a brand name or a person's full name). That isn't officially supported (not sure why), but it's so essential that it was worth hacking around that limitation.
More significant issues: The transcription quality is comparable to other services I've used, but still nowhere near good enough to transcribe our podcasts reliably without a LOT of editing. The "keywords" feature can help a little, but it's not perfect. I also had to jump through many hoops to use entire phrases as keywords (such as a brand name or a person's full name). That isn't officially supported (not sure why), but it's so essential that it was worth hacking around that limitation.
What problems is the product solving and how is that benefiting you?
We're using Deepgram to transcribe podcast episodes for an online publication automatically. Right now, the transcription quality isn't good enough for us to publicize them, but hopefully they're helpful for episode SEO and possibly for accessibility. We don't have the resources to have a human edit each transcript, as we're putting out 10+ episodes per week.
Had great support from the engineering team
What do you like best about the product?
The knowledge of the technical team and their willingness to work with our analyst
What do you dislike about the product?
Confusion over what functionality will be available and when
What problems is the product solving and how is that benefiting you?
It is allowing the bulk transcription of conversation
Great documentation, generous amount of credits and a fast, high quality service.
What do you like best about the product?
I love the design of the API and the related documentation and think the free credits and incentives to earn more by following the tutorials really effective.
What do you dislike about the product?
Diarization is not great – quite a few out-by-one errors related to timings. There are few, if any clientside JavaScript examples in the documentation.
What problems is the product solving and how is that benefiting you?
It provides high-quality transcription at a competitive price, with great performance. The generous amount of free credits you receive helps me test and develop.
My Personal Experience
What do you like best about the product?
The best thing about Deepgram is that it provides code snippets that helps you a lot while building your hobby project or integrating in real world solution. Last year when I created a Voice Search for hackathon then Deepgram's docs and examples helped me a lot in implementing.
What do you dislike about the product?
The one major downside I faced was not a good response from the model. It was working fine with normal language words example. I, we, help etc. But I was looking for a solution that can understand technical words related to programming for example. Java, Python. Although It was working, but I lost top-notch user experience.
What problems is the product solving and how is that benefiting you?
Deepgram is making easy in Creating powerful Voice recognition system for applications.
Great Features
What do you like best about the product?
They have many great features that are easy to turn on or off as your use case for specific sources change. For example, I can choose to turn diarization on for multiple speakers, and I can choose to use individual timestamped words, sentences, and paragraphs and spare myself the trouble of splitting or combining transcripts on my end.
What do you dislike about the product?
Supported input codecs aren't listed on their website, but their salespeople are very helpful via e-mail in this regard.
What problems is the product solving and how is that benefiting you?
I work with large amounts of recorded content, most of it not necessary for the project, and using Deepgram transcripts helps us do keyword searches or skim dialogue to look for what we need.
Next generación voice to text
What do you like best about the product?
Impressive super fast voice to text engine, very accurate thanks to the IA used, improve a lot of features vs other engines, like they Smart format, real time and punctuate
What do you dislike about the product?
Nothing to Say, I think all the córners are covered bye deepgram
What problems is the product solving and how is that benefiting you?
Wey can detect financial frauds, automatizating compliance process, and the QA process to. Wey can use the time saved in other process
One of the more feature-filled and performant transcription solutions on the market!
What do you like best about the product?
- Best-in-class latency for real-time transcription: sends updates quick
- Willing to work with you on building new features, training custom models, and improving the behavior of existing features (for example realtime silence detection was something we really needed)
- Competitive price
- Relatively nice and clean APIs and DevX
- Self-hosting is easy
- Willing to work with you on building new features, training custom models, and improving the behavior of existing features (for example realtime silence detection was something we really needed)
- Competitive price
- Relatively nice and clean APIs and DevX
- Self-hosting is easy
What do you dislike about the product?
- Speaker diarization is not as accurate as other providers (for example Rev.ai)
- Occasional bugs with transcription
- Lots of hidden options & documentation could be improved
- Occasional bugs with transcription
- Lots of hidden options & documentation could be improved
What problems is the product solving and how is that benefiting you?
We use transcription to transcribe conversations and customer calls in real-time on our virtual salesfloor. Deepgram lets us connect to customer calls faster and extract insights from them as they happen for others to see.
Easy to use, fast, powerfull
What do you like best about the product?
The community, the quality of the transcript, the language support
What do you dislike about the product?
Number Spanish support, functionality, can be able more features as very system, text to speech
What problems is the product solving and how is that benefiting you?
transcription system in some different format files
I appreciate Deepgram for its free quotas and reasonable transcription quality
What do you like best about the product?
The large amount of free minutes includes even without purchasing a licence.
The easy-to-use Typescript API, which allowed me to batch-transcribe many audio files at once.
The easy-to-use Typescript API, which allowed me to batch-transcribe many audio files at once.
What do you dislike about the product?
The transcription quality is not always very good, especially for languages outside of English (French for instance). The word whitelisting is sometimes either ignored, or used too much (i.e. Deepgram wrongly detects words in the audio file as being one of those in the whitelist)
What problems is the product solving and how is that benefiting you?
I use Deepgram to generate a quick transcription of podcast contents as subtitle files. After that I go over the transcribed subtitles and fix the issues manually.
showing 41 - 50