Sold by: Smallest AI
Deployed on AWS
bidirectional-WebSocket streaming; diarization and PII redaction built in; fully network-isolated so audio never leaves the account.
Overview
Pulse by Smallest AI is a low-latency, production-ready speech-to-text model built for real-time voice applications and high-volume transcription. It supports real-time WebSocket streaming and pre-recorded batch transcription across 30+ languages, with automatic language detection and code-switching. Pulse includes speaker diarization, word and sentence-level timestamps, keyword boosting, noise-robust transcription, punctuation, and built-in PII/PCI redaction. Its streaming pipeline is optimized for conversational AI, live calls, contact centers, and other latency-sensitive workloads.
Highlights
- Low-latency speech-to-text for real-time transcription on live calls and voice agents (sub-100ms time-to-first-token). Multilingual speech-to-text across 21 streaming languages Built for conversational real-time transcription: partial transcripts, word-level timestamps, and speaker diarization.
Details
Sold by
Categories
Delivery method
Latest version
Deployed on AWS
New
Introducing multi-product solutions
You can now purchase comprehensive solutions tailored to use cases and industries.
Features and programs
Financing for AWS Marketplace purchases
AWS Marketplace now accepts line of credit payments through the PNC Vendor Finance program. This program is available to select AWS customers in the US, excluding NV, NC, ND, TN, & VT.
Pricing
Pricing is based on actual usage, with charges varying according to how much you consume. Subscriptions have no end date and may be canceled any time.
Additional AWS infrastructure costs may apply. Use the AWS Pricing Calculator to estimate your infrastructure costs.
Dimension | Description | Cost/host/hour |
|---|---|---|
ml.g5.2xlarge Inference (Batch) Recommended | Model inference on the ml.g5.2xlarge instance type, batch mode | $18.00 |
ml.g6.2xlarge Inference (Real-Time) Recommended | Model inference on the ml.g6.2xlarge instance type, real-time mode | $18.00 |
ml.g5.xlarge Inference (Batch) | Model inference on the ml.g5.xlarge instance type, batch mode | $18.00 |
ml.g6.4xlarge Inference (Real-Time) | Model inference on the ml.g6.4xlarge instance type, real-time mode | $18.00 |
Dimensions summary
You pay by the hour for the instance that runs model inference. Pricing splits two ways: the processing mode and the instance size. Two batch options run on ml.g5.xlarge and ml.g5.2xlarge for processing recorded audio. Two real-time options run on ml.g6.2xlarge and ml.g6.4xlarge for streaming transcription. Within each mode, the larger instance size carries more compute capacity. You choose the mode that fits your workload, then pick the instance size. Billing is per host hour, so cost scales with how long each instance runs.
Top-of-mind questions for buyers
You pay for the running time of the AWS instance that hosts the model. Each host hour covers one instance kept active, regardless of how much audio you transcribe during that hour. Cost tracks how long the instance stays running, not the number of minutes processed.
Batch options on ml.g5 instances suit recorded audio processed in bulk. Real-time options on ml.g6 instances suit live streaming transcription that needs fast turn-taking. Both meter per host hour. You pick the mode that matches your workload, then choose the instance size for that mode.
Host hour charges apply only while an instance runs. A fully stopped instance accrues no software charges. Underlying AWS storage or related resource fees may still apply depending on your setup. The listing software meters running time, so shutting instances down when unused stops those charges.
www.smallest.ai
Helpful?
Vendor refund policy
N/A
Legal
Vendor terms and conditions
Upon subscribing to this product, you must acknowledge and agree to the terms and conditions outlined in the vendor's End User License Agreement (EULA) .
Content disclaimer
Vendors are responsible for their product descriptions and other product content. AWS does not warrant that vendors' product descriptions or other product content are accurate, complete, reliable, current, or error-free.
Delivery details
Amazon SageMaker model
An Amazon SageMaker model package is a pre-trained machine learning model ready to use without additional training. Use the model package to create a model on Amazon SageMaker for real-time inference or batch processing. Amazon SageMaker is a fully managed platform for building, training, and deploying machine learning models at scale.
Deploy the model on Amazon SageMaker AI using the following options:
Version release notes
Seven European languages, from one checkpoint.
- French, Spanish, Portuguese, Italian, Russian, German and Dutch, selected per request with the language parameter or detected automatically from the audio.
- Live streaming over a bidirectional WebSocket, with partial transcripts returned while the speaker is still talking, and whole-file transcription on POST /invocations.
- Word-level timestamps, keyword boosting, punctuation, profanity filtering, and built-in PII/PCI redaction.
- Runs with network isolation enabled: no outbound network access, so your audio stays inside your AWS account.
Additional details
Inputs
- Summary
audio/wav, audio/x-raw, application/octet-stream or application/json; fr, es, pt, it, ru, de, nl
- Input MIME type
- audio/wav, audio/x-raw, application/octet-stream, application/json
https://docs.smallest.ai/models/api-reference/speech-to-text/speech-to-text
https://docs.smallest.ai/models/api-reference/speech-to-text/speech-to-text
Support
Vendor support
AWS infrastructure support
AWS Support is a one-on-one, fast-response support channel that is staffed 24x7x365 with experienced and technical support engineers. The service helps customers of all sizes and technical abilities to successfully utilize the products and features provided by Amazon Web Services.
Similar products
Deepgram, Language AI models to power your apps.
Deepgram is the leading voice AI platform for enterprise use cases, offering speech-to-text (STT), text-to-speech (TTS), and full speech-to-speech (STS) capabilities. 200,000+ developers build with Deepgrams voice-native foundational models due to our unmatched accuracy, low latency, and pricing. Having processed over 50,000 years of audio and transcribed over 1 trillion words, there is no organization in the world that understands voice better than Deepgram.
Deepgram is the leading voice AI platform for enterprise use cases, offering speech-to-text (STT), text-to-speech (TTS), and full speech-to-speech (STS) capabilities. 200,000+ developers build with Deepgrams voice-native foundational models due to our unmatched accuracy, low latency, and pricing. Having processed over 50,000 years of audio and transcribed over 1 trillion words, there is no organization in the world that understands voice better than Deepgram.
Combine the power of Gen AI with the best of human voice recognition. SoftServe Speech Recognition Platform, accelerated by NVIDIA AI Blueprints and NVIDIA® Riva, converts speech to text with high precision. Unlike other systems, it grasps the nuances of children's voices, making it easier for teachers and other professionals to help kids improve their speech and treatment outcomes. The solution captures even the subtlest vocal nuances, down to the phoneme level. Giving you the ability to gain consistently accurate transcriptions — even in challenging, noisy environments.
Customer reviews
No customer reviews yet
Be the first to review this product . We've partnered with PeerSpot to gather customer feedback. You can share your experience by writing or recording a review, or scheduling a call with a PeerSpot analyst.