gemini-3.5-transcribe-live
Neural Network
Max answer length
(in tokens)
Context size
(in tokens)
Prompt cost
(per 1M tokens)
Answer cost
(per 1M tokens)
Image prompt
(per 1K tokens)
How it works gemini-3.5-transcribe-live?
Frequently asked questions about gemini-3.5-transcribe-live
You can use generated results for commercial purposes. You own all rights to the content you create. The only restriction: make sure your prompt does not include copyrighted third-party material. You are responsible for respecting the rights to any input data.
gemini-3.5-transcribe-live is a model from google-gemini that, according to our data, generates audio: you describe the task in text, and you get an audio track as output. It is suitable for voiceovers for videos, podcasts, educational materials, and voice scenarios in products. Access from Russia without a VPN or foreign card, pay-as-you-go for tokens.
Specific formats and bitrates are not specified in our data, so rely on the result in the interface: the audio can be listened to and downloaded. A context of 65,536 tokens and up to 32,000 tokens per response allows working with long texts in a single request.
There is no note about reasoning before answering in our data, and this is not a denial of the possibility, but a lack of confirmation. For voiceovers, detailed reasoning is usually not needed: it is more important to formulate the task accurately. If you need long chains of reasoning, switch to a text model in the same window.
Function calling and structured JSON output are not noted in our data, so do not build them into your architecture without verification. Access is via the unified OpenAI-compatible BotHub API, so it is convenient to offload tool usage to a text model without rewriting the integration.
File acceptance beyond text is not noted in our data, so count on a text description of the task. The 65,536 token context is enough for a voluminous script in its entirety. If you need analysis of an image, recording, or video, select a model with the appropriate input in the same window.
The language of prompts and voiceovers is not fixed in our data, so we will not promise anything — it is easier to check on a short fragment and compare with another model in the same window. The BotHub interface, support, and payment with Russian cards are available without a VPN or foreign card.