OPEN SPEECH MODELS

Salesforce releases SOV-Transcribe-1 for configurable recorded-audio transcription

By IDAKIM Digital · Published November 13, 2026

Apache-licensed weights offer self-hosting and instruction-guided output, with an emphasis on difficult speech and named entities.

Salesforce’s Sovereign Intelligence team published SOV-Transcribe-1 on November 12. The model targets non-streaming automatic speech recognition, including terminology, accents and noisy recordings. Its release describes instruction-guided transcript formatting and contextual keywords, allowing an application to specify more than a plain verbatim transcript. The announcement states four billion active parameters in a mixture-of-experts architecture and makes weights available under Apache 2.0. Salesforce reports strong results against open-weight transcription models, but these are release-time comparisons rather than validation of a reader’s call recordings. Self-hosting can give operators more control over audio processing, while introducing hardware, deployment and retention responsibilities. The release should not be treated as a drop-in streaming conversation engine: batch accuracy, entity capture and interactive latency are different requirements.

Why it matters

Evaluate consented recordings with names, numbers and domain vocabulary. Check transcription errors and formatting separately, then size the serving hardware before selecting self-hosting.

Source: Salesforce Sovereign Intelligence · November 12, 2026

Product claims are attributed to the linked publisher. IDAKIM Digital has not independently tested these claims.