Sold by: Sigmodata
Deployed on AWS
Free Trial
Whisper speech-to-text as a SageMaker model package. Send a WAV, MP3, or JSON (base64) payload and receive a transcript. Runs on CPU; uses a GPU automatically when you deploy on a GPU instance.
Overview
Image
Python predict
Image
POST /invocations
This SageMaker model package transcribes English speech using whisper.cpp (baked base.en weights). Deploy a real-time endpoint or run SageMaker batch transform. Send JSON with a base64-encoded audio field, or raw audio bytes with Content-Type audio/wav, audio/mpeg, or application/octet-stream. The response is JSON with a text field.
GPU is optional. Deploy on a GPU instance (for example ml.g4dn.xlarge) for higher throughput; the same image uses CUDA automatically when a GPU is present, and falls back to CPU otherwise. Real-time payloads are limited to 6 MB; longer files should use batch transform.
This listing is the SageMaker path. For a self-hosted HTTP API, web UI, and selectable Whisper models, use the Speech-to-Text Server AMI or container.
Common use cases include transcribing call-centre recordings, meeting and interview audio, podcast and media back-catalogues, and voice notes, and building private speech analytics pipelines where audio must stay inside your own AWS account.
Model and training data The model is OpenAI's Whisper base.en, an English-only encoder-decoder speech recognition model, served through whisper.cpp with the weights baked into the image. Sigmodata did not train it; it was trained by OpenAI on a large corpus of web audio and is used here under the MIT licence. base.en is the small end of the Whisper family, chosen so the model runs acceptably on CPU.
Measured performance
- Word error rate 4.4 percent on LibriSpeech test-clean, measured over 300 utterances and 6,428 reference words. Scoring lowercases, strips punctuation other than apostrophes, and collapses whitespace.
- Measured by sending the audio to the same container image the model package ships, through the same /invocations endpoint a buyer calls.
- LibriSpeech test-clean is read speech recorded in good conditions. Expect a higher error rate on spontaneous conversation, telephony audio, background noise, or strong accents.
Known limitations
- English only. Other languages are not supported; use a multilingual Whisper model if you need them.
- base.en is the smallest English Whisper model. Larger models are more accurate; the Speech-to-Text Server AMI and container let you pick one.
- The response is plain text. Word-level timestamps, speaker diarisation, and segment metadata are not returned by this endpoint.
- Real-time invocations accept up to 6 MB. Use batch transform for longer audio; JSON base64 inflates a file by roughly one third.
- Accuracy degrades on overlapping speech and on very short clips with little context.
We welcome your feedback at aws-support@sigmodata.com . Sample notebook: https://www.sigmodata.com/products/speech-to-text-model/getting-started.ipynb
Highlights
- Transcribe English audio on a SageMaker endpoint - WAV, MP3, or JSON base64
- Same CUDA image as the Speech-to-Text Server, with automatic CPU fallback
- Real-time endpoints and batch transform; GPU if you have one, CPU if you don't
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
Free trial
Try this product free for 5 days according to the free trial terms set by the vendor.
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.xlarge Inference (Real-Time) Recommended | Model inference on the ml.g5.xlarge instance type, real-time mode | $0.20 |
ml.m5.xlarge Inference (Batch) Recommended | Model inference on the ml.m5.xlarge instance type, batch mode | $0.20 |
ml.c5.xlarge Inference (Batch) | Model inference on the ml.c5.xlarge instance type, batch mode | $0.20 |
ml.c5.xlarge Inference (Real-Time) | Model inference on the ml.c5.xlarge instance type, real-time mode | $0.20 |
ml.g4dn.xlarge Inference (Batch) | Model inference on the ml.g4dn.xlarge instance type, batch mode | $0.20 |
ml.g4dn.xlarge Inference (Real-Time) | Model inference on the ml.g4dn.xlarge instance type, real-time mode | $0.20 |
ml.g5.xlarge Inference (Batch) | Model inference on the ml.g5.xlarge instance type, batch mode | $0.20 |
ml.m5.large Inference (Batch) | Model inference on the ml.m5.large instance type, batch mode | $0.20 |
ml.m5.large Inference (Real-Time) | Model inference on the ml.m5.large instance type, real-time mode | $0.20 |
ml.m5.xlarge Inference (Real-Time) | Model inference on the ml.m5.xlarge instance type, real-time mode | $0.20 |
Dimensions summary
You pay by the hour for each hosting instance you run. Pricing splits across five instance types: two general-purpose CPU options (ml.m5.large, ml.m5.xlarge), one compute-optimized CPU option (ml.c5.xlarge), and two GPU options (ml.g4dn.xlarge, ml.g5.xlarge). Each instance type offers two modes. Batch mode processes longer audio files through batch transform. Real-time mode serves live endpoint requests. Your cost scales with the instance size you choose, whether you use a GPU, the mode you run, and total hours running. GPU instances raise throughput; CPU instances run without a GPU.
Top-of-mind questions for buyers
Real-time mode serves live requests through a hosted endpoint. It accepts payloads up to 6 MB per invocation. Batch mode uses batch transform to process longer or larger audio files at once. Choose real-time for short clips and live calls; choose batch for long recordings.
A host-hour is one hour of running the SageMaker instance you selected, for one instance type in one mode. You pay for the time the endpoint or batch job stays running. Charges stop when the instance is not running. Multiple running instances each accrue their own host-hours.
Each instance type has its own hourly rate, so your cost depends on the type you deploy. The same container image runs on both. GPU instances like ml.g4dn.xlarge and ml.g5.xlarge raise throughput. CPU instances like ml.m5.large and ml.c5.xlarge run without a GPU using CPU fallback.
www.sigmodata.com
Helpful?
Vendor refund policy
No refunds offered but you may cancel at any time
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
Updated listing text. No change to the API or the model.
Additional details
Inputs
- Summary
JSON with a base64 "audio" string and optional "filename", or raw audio bytes (WAV/MP3) for real-time and batch transform. Max 6 MB per real-time request.
- Input MIME type
- audio/wav, audio/mpeg, application/octet-stream, application/json
{"audio": "<base64 WAV or MP3>", "filename": "clip.wav"}
(raw WAV or MP3 bytes, one file per record)
Input data descriptions
The following table describes supported input data fields for real-time inference and batch transform.
Field name | Description | Constraints | Required |
|---|---|---|---|
audio | Base64-encoded WAV or MP3 (JSON Content-Type only) | - | No |
filename | Optional filename hint for the decoder (for example clip.wav) | - | No |
Resources
Vendor resources
Support
Vendor support
Support contact: Email: aws-support@sigmodata.com Support URL: https://www.sigmodata.com
Support description: Support via email for deployment, configuration, and troubleshooting.
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.
Natural and expressive voices in multiple languages. For voice agents and brand ambassadors.
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.