At Speechly, we are known for the speed of our Automatic Speech Recognition and Natural Language Understanding. This is all thanks to our Streaming API.
However, as our user base has continued to grow so has the demand for new product features. One of the most requested features to date is the ability to use Speechly for transcribing large amounts of pre-recorded audio or video content.
Given that demand, we are excited to release the Speechly Batch API for Enterprise users.
Transcribe Large Amounts of Pre-Recorded Audio and Video with the Speechly Batch API
The Speechly Batch API enables users to easily and privately send large sets of pre-recorded audio or video files to Speechly for Speech Recognition. This makes it easy to complete tasks like Transcription, Moderation or other types of Speech Analysis on large amounts of off-line audio or video data.
To use the Speechly Batch API, users simply submit the audio and read the results after it is processed. The throughput of the Speechly Batch API supports processing thousands of hours of audio per hour. You can submit the audio directly to Speechly or give the Google Storage or Amazon S3 file URL, making it easy to scale.
With the Speechly Batch API you can expect the same performance and accuracy that you would expect from the Streaming API. We also offer Enterprise customers data annotation services for continued monitoring and improvement in the performance of your Speech Recognition.
When deploying the Speechly Batch API, customers have the ability to deploy On-Premise or in a Private Cloud. These deployment options help customers leverage Speech Recognition in the most secure and cost-effective manner possible.
Example Scenarios for the Speechly Batch API
The use cases for the Speechly Streaming API and the Speechly Batch API are the same, however the Streaming API is used for live online scenarios while the Batch API is for offline scenarios. This means you should use the Streaming API if you need to process speech in real-time as it’s being recorded. Use the Batch API if your audio or video has already been recorded and you need to process later on. Below are a few examples where the Speechly Batch API is ideal:
User Generated Content Monitoring - Massive amounts of podcasts and video are being recorded and uploaded everyday. Speechly makes it easy to transcribe any user generated audio or video content uploaded to your platform for use cases such as Moderation, Content Categorization or Content Indexing, Measuring Brand Awareness and Strategic Ad Placement.
Meeting Analysis - The Speechly Batch API makes it simple to convert meetings into transcriptions for later analysis. This can be used for use cases such as creating Meeting Summaries or making it easy to search back through past discussions for specific information.
Customer Support - Speechly can quickly and accurately transcribe recorded customer calls. This can be used for use cases such as Measuring Agent Performance and Agent Training or extracting & documenting relevant customer support information. This is different from monitoring a conversation in Real-Time, where the goal could be to offer the Agent relevant information as the call is happening. This would be another example of where the Speechly Streaming API would be better suited.
If you are interested in learning more about how to use the Speechly Batch API, you can read more in our Documentation. If you would like to learn more about getting access to the Speechly Batch API, Contact Us today.