Google introduces new speech model

Google announced the release of a new speech-to-text model called Gemini 3.5 Transcribe, which is considered the most accurate the company has developed so far in this field. The new model offers greater contextual understanding with high precision and supports automatic recognition of over 85 languages. It will replace the previous model, "Chirp 3."

Size: 120%Text Size

Notice: This Is an AI-Generated Summary

Display The Full Article