audioTimestamp option missing in GenerationConfig

[jclab-joseph]

Description of the feature request:

See https://cloud.google.com/vertex-ai/generative-ai/docs/multimodal/audio-understanding?hl=ko#:~:text=files%2C%20enable%20the-,audioTimestamp,-parameter%20in%20GenerationConfig

But I am not able to set this parameter in GenerationConfig.

What problem are you trying to solve with this feature?

Ensures that timestamps are output accurately when transcribing audio.

Any other information you'd like to share?

No response

[hkt74]

Hi @jclab-joseph

Thanks for bringing this to our attention! Just to clarify, the API documentation you linked is for Vertex AI (on GCP). You can find code samples for Vertex AI here: https://cloud.google.com/vertex-ai/generative-ai/docs/reference/nodejs/latest

This SDK is specifically for the Gemini Developer API https://ai.google.dev/gemini-api/docs, which unfortunately doesn't currently support the audioTimestampconfiguration option.

[PaulMest]

audioTimestamp is now available in their genai SDK which works across Vertex AI and AI Studio.
source