Sold by: pyannoteAI
Deployed on AWS
Accurately identify who spoke when in any audio using our Speaker Diarization. This scalable, language-agnostic service segments multi-speaker audio into speaker-labeled time intervals, supporting formats like WAV, MP3, and FLAC. Ideal for transcription, call analytics, media processing, and compliance workflows.
Overview
The Speaker Diarization API enables accurate segmentation of audio recordings by detecting and labeling individual speakers across time. Designed for seamless integration into transcription pipelines, media workflows, and audio analytics systems, it supports a wide range of formats including WAV, MP3, FLAC, and OGG. The service is language-agnostic and works across diverse audio sourcecalls, meetings, interviews, podcasts, and more. With built-in support for mono and stereo channels, varying sample rates, and flexible input options it can be deployed in batch or near-real-time use cases. Key features include automatic speaker count estimation, precise time-stamped speaker labeling, and detection of overlapping speech. Outputs are returned in structured JSON for easy integration with transcription engines, search indexes, or business intelligence tools. Whether you are enriching speech-to-text transcripts, analyzing call center performance, or processing long-form media, this API improves clarity, organization, and data usability.
Highlights
- Accurate speaker diarization for multi-speaker audio, with support for automatic speaker count estimation and overlapping speech detection.
- Language-agnostic and format-flexible: Works with WAV, MP3, FLAC, and more; supports mono and stereo channels for diverse use cases like transcription and media analysis.
Details
Sold by
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.g4dn.xlarge Inference (Batch) Recommended | Model inference on the ml.g4dn.xlarge instance type, batch mode | $3.404 |
ml.g4dn.xlarge Inference (Real-Time) Recommended | Model inference on the ml.g4dn.xlarge instance type, real-time mode | $2.714 |
Dimensions summary
You pay by the hour for model inference running on a single instance type, the ml.g4dn.xlarge. The two dimensions split by processing mode. Batch mode handles pre-recorded audio files submitted for diarization, identifying who spoke when. Real-time mode processes streaming audio as a conversation unfolds, delivering speaker labels with low latency. Both bill per host hour, so your cost scales with how long each instance runs. Choose the mode that matches your workload: batch for recorded files, real-time for live streaming pipelines.
Top-of-mind questions for buyers
One host hour is one hour that a single ml.g4dn.xlarge instance runs the diarization model. Billing counts wall-clock running time per instance, not the length of audio processed. If you run multiple instances at once, each accrues its own host hours in parallel.
Both modes bill per host hour on the same instance type. Batch mode processes pre-recorded audio files submitted as jobs. Real-time mode processes streaming audio live over a connection, returning speaker labels with low latency. Your cost tracks how long the instance runs, whichever mode you choose.
Charges accrue per host hour while the instance runs, whether or not it is actively processing audio. Idle time on a running instance still meters host hours. To stop software charges, stop the instance. Underlying AWS infrastructure fees may still apply for stored resources.
pyannote.ai
Helpful?
Vendor refund policy
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
Precision-2 model with improved diarization capabilities (37% accuracy improvement). New optional min_speakers and max_speakers input arguments.
Additional details
Inputs
- Summary
Diarization input: { "audio": "", "num_speakers": 2 }
- Input MIME type
- application/json
https://github.com/pyannoteai/aws-marketplace-docs/blob/main/example_files/marklex1min.mp3
https://github.com/pyannoteai/aws-marketplace-docs/blob/main/example_files/marklex1min.mp3
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 audio | - | No |
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.
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.