Sarvam AI launches Saaras V4, its most capable speech-to-text model yet
itsOmSarraf_ · x · 2026-09-25
Indian AI company Sarvam AI unveiled Saaras V4, its most capable speech-to-text model yet, delivering strong performance across English and Indian languages with better accuracy over a much wider range of speech. It was built with specific attention to noise, accents, dialects, and mixed-language speech. Details in the official blog.
More from Multimodal
- AI-Made Short Film Adapts Olaf Stapledon's 1937 Novel Star Maker — rafpavon · 2026-09-25
- Open-sourced skill turns your codebase into a polished product promo video via Claude Code — op7418 · 2026-09-25
- Free HY Image3.5 Preview turns creative briefs into polished posters and e-commerce visuals — anthara_ai · 2026-09-25
- faster-qwen3-tts brings quantized Qwen3-TTS to Apple Silicon via GGML — andimarafioti · 2026-09-25
- Shanghai AI Lab's AV-GRPO Uses Modality-Anchored RL to Beat LTX-2.3 at Joint Audio-Video Generation — Shanghai-AI-Laboratory · 2026-09-25
- Blender API and licensing limits make EV Baker add-on best under 20-50M polygons — ssh4net · 2026-09-25