Salesforce's SOV-Transcribe-1 tops open-weights speech-to-text at 2.3% AA-WER
Artificial Analysis ranks Salesforce's SOV-Transcribe-1 as the top open-weights speech-to-text model at 2.3% AA-WER with 4.6B active parameters.
Artificial Analysis ranks Salesforce's SOV-Transcribe-1 as the top open-weights speech-to-text model, with a 2.3% AA-WER score and 4.6B active parameters. Word error rate measures the share of words a transcription model gets wrong, so lower is better. Active parameters are the share of a model's weights used when it processes an input.
The ranking rests on a single leaderboard result. Artificial Analysis is the only source cited, and no other transcription benchmark or independent test is reported.
What changed
Type: Other · Applies to: SOV-Transcribe-1
Why it matters
Teams building transcription pipelines get another open-weight option to benchmark, though the claim rests on one leaderboard.
Original source
We have no link to the original source for this story yet.
Where we saw it
Latent Space
AI wrote this summary from the sources listed, and nobody reviewed it individually before publication. Check the original source before you rely on it. How it works