AI Speech to Text

OCI Speech is an AI service that applies automatic speech recognition technology to transform audio-based content into text. Developers can easily make API calls to integrate OCI Speech’s pretrained models into their applications. OCI Speech can be used for accurate, text-normalized, time-stamped transcription via the console and REST APIs as well as command-line interfaces or SDKs. You can also use OCI Speech in an OCI Data Science notebook session. With OCI Speech, you can filter profanities, get confidence scores for both single words and complete transcriptions, and more.

OCI Speech features

Prebuilt acoustic and language models

OCI Speech uses automatic speech recognition, a deep learning process, to derive accurate transcription from natural conversations. Get started easily by using prebuilt acoustic and language models that don’t require users to have data science experience.

Analyze data from audio and video files

Search, index, and decipher data buried in your audio files. Convert recorded audio conversations to textual data to analyze with AI services. For example, you can use OCI Language to retrieve the sentiment and OCI Speech’s anomaly detection capabilities to identify chances of customer churn.

Native multilingual support

OCI Speech ASR models support English, Spanish, and Portuguese, so you can transcribe your audio files in your preferred languages.

Diarization for increased readability and understanding

OCI Speech supports diarization for organizing, analyzing, and extracting meaningful information from spoken interactions.

Integrated transcription service

Eliminate reliance on third-party transcription offerings and practice more control over your data with end-to-end security and compliance.

Easy to integrate

OCI Speech is a versatile service that can be called via REST APIs, different SDKs, and Oracle CLI. Developers can easily deploy a scalable speech service without having data science or ML expertise.

Purpose-built for security and privacy

Oracle Cloud Infrastructure Speech protects our customers’ privacy. Prebuilt automatic speech recognition models transcribe your content, but do not store any data for training, debugging, or other purposes.

Integrated transcription service

OCI Speech uses proprietary models and architecture that enables fast conversion for speech into text.

Confidence score per word

We added a word-level confidence score to help you identify words that might have been transcribed incorrectly. Use the word confidence score to determine where to focus when building an application.

Profanity filters

We added prebuilt word filtering using a curated list of profanities. You can either mask, remove, or tag profanities.

OCI Speech use cases

  • Customer feedback analytics

  • Digital Media content search and closed captions

    Automatically provide in-workflow closed caption on OCI platform for all content created and curated by digital media service. Index your content using OCI speech for easy search across your content.

  • Call centers, call analytics

    OCI Speech can transcribe customer calls for easy search and retrieval of information. Use OCI Language and Anomaly Detection together to detect sentiment and identify customer churn and staff training opportunities.

OCI Speech resources

April 27, 2022

Punctuation, Closed Captions and 8kHz models are now available for OCI Speech.

Guy Michaeli, Senior Principal Product Manager

Today, we are happy to announce three new capabilities for the Speech service at no additional cost: native support for 8kHz audio files, support for output in the SRT (a closed- caption file format), and automatic punctuation of output text. These new capabilities are now available in all OCI's commercial regions and are part of our commitment to provide high-quality, affordable transcription for our customers.

Read the full article

Get started with OCI Speech

Oracle Cloud Free Tier

Build, test, and deploy applications on Oracle Cloud—for free.