Product roadmap
We ship batch-first because that's what voice AI teams need most today. Here's what's live and what's next.
Available now
Standard batch transcription
High-priority batch transcription with the full feature set included.
Dynamic multilingual detection
Mid-file language switching across all Whisper v3 Turbo languages.
Six output formats
JSON, TXT, SRT, VTT, DOCX, PDF with timestamps and speaker labels.
Coming soon
Economy batch transcription
Idle-compute queue at $0.0015/min. 24-hour SLA with explicit failure reporting.
Real-time streaming transcription
WebSocket-based live transcription for voice agents and meeting apps.
Domain-adaptive models
Automatic routing to specialized LoRA models for medical, legal, and finance — with mid-file domain switching.
Direct URL upload
Transcribe audio from a URL without SDK — REST endpoint for files up to 200 MB.
EU data residency
Process and store audio in EU regions (AWS + OVH).
SOC 2 Type II & HIPAA
Enterprise compliance certifications for regulated industries.