aiqbench takeChips & Infra

This collaboration demonstrates a significant efficiency gain in running high-volume speech-to-text workloads on standard CPUs rather than requiring dedicated GPU accelerators. By optimizing the Whisper model to half its original parameter count, the team doubled throughput while reducing memory usage, offering a viable path for enterprises seeking cost-effective, on-premise AI solutions that prioritize data sovereignty.

Source coveringQuantum Zeitgeist

What this isA model's reading of the article and related coverage, written for aiqbench. Not the publisher's words.

Intel, HPE and Multiverse Computing double speech-to-text speed

Quantum ZeitgeistSeptember 24, 20263 min read

Intel, HPE & Multiverse Computing doubled transcription rates, achieving 17,000 audio hours/day on a single server—all with CPUs, not GPUs.

A single Hewlett Packard Enterprise ProLiant DL380 server now transcribes approximately 17,000 hours of audio daily using only CPUs, a feat previously requiring specialized hardware. Multiverse Computing, in collaboration with HPE and Intel, achieved this by optimizing the Whisper speech-to-text model to 0.4 billion parameters, half its original size, without sacrificing quality, the company says. “Transcription is one of…

Excerpt shown under the publisher's RSS terms · 59 words · full text and figures are on the publisher's site

Read at Quantum Zeitgeist ↗Ask about this article