Exciting Updates in 3.7

Share This Post

We are excited to share the details of our latest platform update, which brings significant enhancements across various domains. 

Enhancing PrivateGPT and UI Experience

Our recent software update introduces significant enhancements, focusing on the PrivateGPT experience and user interface improvements. Here’s a summary of these updates:

Customizable Conversation Roles

We’ve added the functionality to define roles for conversations. Roles are text prompts injected into the beginning of all conversations. This feature allows for more tailored and effective responses, enhancing the overall user experience with PrivateGPT.

Administration and Configuration Ease

We are introducing an administration page in the Customer Portal. This addition is designed for admins who want to configure their PrivateGPT deployments with point-and-click. Take a look at this video overview for more information.

Cohere Support

PrivateGPT now supports Cohere LLM, offering users a broader choice of language models while maintaining our commitment to data privacy. 

UI Enhancements

We have implemented UI improvements as part of our commitment to providing a seamless and intuitive user experience. These enhancements are designed to improve front-end deployment, making the interface more user-friendly and responsive.

Image, Audio and ASR Processing Improvements

Configure Maximum Image Pixels

The container allows users to configure the maximum allowed pixels in images processed using the PAI_MAX_IMAGE_PIXELS environment variable. 

Improved Performance of Standard ASR

We have enhanced the Standard Automatic Speech Recognition (ASR) performance results to provide more accurate and efficient audio data processing.

New Audio Options

The platform now includes enhanced audio file processing capabilities, allowing users to customize bleep frequency and gain through two new parameters: bleep_frequency and bleep_gain. These adjustments are accessible under AudioOptions in the process_file routes, as detailed in our API specification.

Model Improvements Across Languages and Domains

Multi-Language Improvements

Japanese: Enhanced average recall of PII entity types, bolstering the identification and protection of sensitive personal information in Japanese.

Health Data Improvements

Spanish & Portuguese:  Advanced detection capabilities for NAME_MEDICAL_PROFESSIONAL and ORGANIZATION_MEDICAL_FACILITY, essential for accurately identifying personal and organizational information in healthcare contexts in Spanish and Portuguese.

MEDICAL_CODE Beta Support: Introduction of the MEDICAL_CODE entity type in beta, aiding in categorizing various medical classification codes from systems like ICD-10, NDC, and SNOMED. 

Subscribe To Our Newsletter

Sign up for Private AI’s mailing list to stay up to date with more fresh content, upcoming events, company news, and more! 

More To Explore

Download the Free Report

Request an API Key

Fill out the form below and we’ll send you a free API key for 500 calls (approx. 50k words). No commitment, no credit card required!

Language Packs

Expand the categories below to see which languages are included within each language pack.
Note: English capabilities are automatically included within the Enterprise pricing tier. 

French
Spanish
Portuguese

Arabic
Hebrew
Persian (Farsi)
Swahili

French
German
Italian
Portuguese
Russian
Spanish
Ukrainian
Belarusian
Bulgarian
Catalan
Croatian
Czech
Danish
Dutch
Estonian
Finnish
Greek
Hungarian
Icelandic
Latvian
Lithuanian
Luxembourgish
Polish
Romanian
Slovak
Slovenian
Swedish
Turkish

Hindi
Korean
Tagalog
Bengali
Burmese
Indonesian
Khmer
Japanese
Malay
Moldovan
Norwegian (Bokmål)
Punjabi
Tamil
Thai
Vietnamese
Mandarin (simplified)

Arabic
Belarusian
Bengali
Bulgarian
Burmese
Catalan
Croatian
Czech
Danish
Dutch
Estonian
Finnish
French
German
Greek
Hebrew
Hindi
Hungarian
Icelandic
Indonesian
Italian
Japanese
Khmer
Korean
Latvian
Lithuanian
Luxembourgish
Malay
Mandarin (simplified)
Moldovan
Norwegian (Bokmål)
Persian (Farsi)
Polish
Portuguese
Punjabi
Romanian
Russian
Slovak
Slovenian
Spanish
Swahili
Swedish
Tagalog
Tamil
Thai
Turkish
Ukrainian
Vietnamese

Rappel

Testé sur un ensemble de données composé de données conversationnelles désordonnées contenant des informations de santé sensibles. Téléchargez notre livre blanc pour plus de détails, ainsi que nos performances en termes d’exactitude et de score F1, ou contactez-nous pour obtenir une copie du code d’évaluation.

99.5%+ Accuracy

Number quoted is the number of PII words missed as a fraction of total number of words. Computed on a 268 thousand word internal test dataset, comprising data from over 50 different sources, including web scrapes, emails and ASR transcripts.

Please contact us for a copy of the code used to compute these metrics, try it yourself here, or download our whitepaper.