Acrebase tested 11 Speech to Text models for WER, CER, and silence hallucination.
Public source
Publisher name
Public post
This surprised us! Acrebase tested 11 Speech to Text models for WER (Word Error Rate), CER (Character Error Rate) and silence hallucination. GPT-4o mini Transcribe perfo…
Company
Acrebase
Smarter negotiation. Built for Commercial Real Estate.
- Company size
- 2–10 employees
About Acrebase
Smarter negotiation. Built for Commercial Real Estate.
See moreDiscover more
Similar signals
Similar public activity from other companies.
Research & Knowledge
Speechmatics
Speechmatics benchmarked Melia 1 for switch-point accuracy across Arabic, Mandarin, and Tamil paired with English against Soniox, Amazon, and AssemblyAI.
Research & Knowledge
Besimple AI
Besimple AI published how Thinking Machines Lab's Inkling speech model was fine-tuned on 1, 25, and 100 hours of proprietary data containing alphanumeric entities.
Research & Knowledge
AudioShake
AudioShake transcribing separated tracks produces 4.1× fewer transcription errors against the strongest open-source speech separator.
Research & Knowledge
Coval
Coval published benchmarks showing AssemblyAI Universal 3.5 Pro pulls ahead on accuracy at 3.0% WER.
Research & Knowledge
Shunya Labs