V‑Blaze REST API version 2.1.0 is a major release with numerous improvements.

1. API-based ASR flows demonstrate more than 30% improved throughput performance for typical call audio.

2. V‑Blaze now supports chunked transfer encoding in POSTs. In chunked transfer encoding, the data stream is divided into a series of non-overlapping "chunks." The chunks are sent out and received independently of one another. Chunked encoding is useful when sending larger amounts of data.

3. V‑Blaze now supports HTTP basic authentication for callbacks. When a callback server requires authentication information, prepend your access credentials to the hostname in the URL as shown in the following example:

   https://username:password@hostname.com

4. Added two new options for /transcribe requests when using utterance_callback :

   1. output=noutts - Returns basic details of the transcript with utterances excluded.

   2. utterance_fmt= OPTION - Controls the output type that would be sent back on a callback. Options are json (default), jsontop , text .

5. Added elements to /statusfull JSON output:

   1. memlow

      Indicates the lowest reading of available memory observed since the last (re)start of the V‑Blaze server.

   2. memhuge

      Provides insight into fragmentation of free memory by indicating the total size of free memory in the NORMAL zone.

      Free memory is counted in huge fragments. In this context, huge is defined as either 512*pagesize or 1024*pagesize. The average pagesize is 4KiB, so the size of a huge fragment could range from 512*4KiB to 1024*4KiB.

   3. maxdecoders

      Indicates the maximum number of audio decoders that can be used by V‑Blaze when processing audio files. Each decoder transcribes a single utterance at a time.

   Refer to the statusfull page for an example that includes these additions.