Sold by: Trellis Data
Deployed on AWS
Free Trial
Leverage Trellis Data's expertise in fine-tuning AI for translation and transcription. Our specialists design and optimize models to handle bespoke languages and dialects, delivering accurate, context-aware results where off-the-shelf tools fall short.
Overview
Trellis Data addresses the limitations of off-the-shelf translation and transcription tools by developing custom AI solutions capable of understanding and processing bespoke languages.
We create production ready models that bridge the gap where mainstream options fall short, equipping enterprises, government agencies, and high-security organizations with language solutions tailored to their unique needs and use cases. We specialize in training, fine-tuning, and optimizing speech-to-text AI models, ensuring accurate output for niche, low-resource, or specialized languages. Our models are designed to deliver optimal results, while providing the linguistic accuracy and cultural sensitivity required for mission-critical applications. Our models are already deployed and trusted by law enforcement and high security agencies in government.
Our transcription models run much faster than real time (depending on the underlying GPU hardware you choose); this means the price per hour of processed audio is much lower than the price to run the model for an hour. We support batched and realtime operation so you can optimize appropriately to your needs.
Highlights
- Accurate transcription with preprocessing aimed at handling a variety of audio sources.
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.g4dn.xlarge Inference (Real-Time) Recommended | Model inference on the ml.g4dn.xlarge instance type, real-time mode | $2.39 |
ml.m5.xlarge Inference (Batch) Recommended | Model inference on the ml.m5.xlarge instance type, batch mode | $1.59 |
ml.g5.xlarge Inference (Batch) | Model inference on the ml.g5.xlarge instance type, batch mode | $3.03 |
ml.g5.xlarge Inference (Real-Time) | Model inference on the ml.g5.xlarge instance type, real-time mode | $3.03 |
ml.g6.xlarge Inference (Real-Time) | Model inference on the ml.g6.xlarge instance type, real-time mode | $1.59 |
ml.m5.xlarge Inference (Real-Time) | Model inference on the ml.m5.xlarge instance type, real-time mode | $1.59 |
Dimensions summary
You pay by the hour for each host that runs the English speech transcription model. Pricing is organized by two choices. First, you pick an AWS instance type: ml.g4dn.xlarge, ml.g5.xlarge, ml.g6.xlarge, or ml.m5.xlarge. These differ in compute power, so hourly rates vary. Second, you pick a processing mode. Real-time mode transcribes as audio arrives and is available on all four instance types. Batch mode processes stored audio in groups and is available on ml.g5.xlarge and ml.m5.xlarge. Costs scale with how many hours each host runs.
Top-of-mind questions for buyers
You pay for each host instance while it runs the transcription model, measured in host hours. Billing meters running time. A host that is fully stopped does not accrue software charges. Underlying AWS storage or other resource fees may still apply while an instance is stopped.
Real-time mode transcribes audio as it streams in, so the host runs continuously while you listen. Batch mode processes stored audio files in groups, so the host runs only during processing jobs. Real-time suits live audio; batch suits recorded files processed later.
Your bill scales with host hours, not the amount of audio itself. The instance type sets the hourly rate, and the number of hours each host runs sets total cost. Running more hosts or longer sessions raises charges. Each running host meters independently.
trellisdata.com
Helpful?
Vendor refund policy
Refund Policy (Effective: 09/10/2026) Refunds for Trellis Data Speech Transcription models may be granted for technical issues unresolvable by support, billing errors, or duplicate charges. Services already rendered, custom configurations, or costs from customer misuse or misconfiguration are non-refundable. This policy operates per AWS Marketplace terms. Contact: support@trellisdata.com.au (9AM-5PM AEDT, weekdays). Policy subject to updates.
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
- Support for Sagemaker Invocations
Additional details
Inputs
- Summary
Supports file inputs with the following media types ['audio/wav', 'audio/mpeg']. For files larger than 30 seconds the model will split the audio into 30 second chunks for processing and provide the output per chunk.
Please include the content-type when sending a file for inference. See sample notebooks linked.
- Limitations for input type
- Files limited to the MIME types specified.
https://github.com/Trellis-Data/sagemaker-samples/tree/main/asr/data/input/real-time
https://github.com/Trellis-Data/sagemaker-samples/tree/main/asr/data/input/batch
Input data descriptions
The following table describes supported input data fields for real-time inference and batch transform.
Field name | Description | Constraints | Required |
|---|---|---|---|
ContentType | Please specify a valid content type matching the file being uploaded. One of ['audio/wav', 'audio/mpeg'] | - | No |
Support
Vendor support
Email our team at: support@trellisdata.com.au
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.
Parakeet-tdt-0.6b-v2 is a 600-million-parameter automatic speech recognition (ASR) model designed for high-quality English transcription, featuring support for punctuation, capitalization, and accurate timestamp prediction.
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.