Initiated by the University of Michigan Herbarium, VoucherVision harnesses the power of large language models (LLMs) to transform the transcription process of natural history specimen labels.
Currently, the cost of OCR API access, such as Google Vision aren't included in the cost calculations, expense report, or cost estimations. They are inexpensive, but as LLM costs go down, they are becoming a larger portion. Google Vision API is now about 10x more per specimen than Gemini 1.5 Flash.
Currently, the cost of OCR API access, such as Google Vision aren't included in the cost calculations, expense report, or cost estimations. They are inexpensive, but as LLM costs go down, they are becoming a larger portion. Google Vision API is now about 10x more per specimen than Gemini 1.5 Flash.