OpenAI · Transcription
GPT Live Transcribe
OpenAI's low-latency streaming speech-to-text model. Served only through the gateway's /v1/realtime WebSocket endpoint as a transcription session (intent=transcription) or as the input-audio transcription model of a speech-to-speech session, billed per minute of audio.
- Maker's price
- $0.017 /min
- Max context
- 32K
- Listings
- 7
- Providers
- 5
Evidence and sources
Fair value (48.6). Costs per task about what its benchmark results predict. From 1 benchmark with measured task costs. Little data behind it, so treat it as a rough estimate.How Value for money worksMaker HQ: US (according to OpenAI's own website)
Sources: OpenAI API data, FlowDot data, LiteLLM data, Models.dev data, LLMGateway dataPrimary data from OpenAI API
from FlowDot: name
from LiteLLM: variant, deprecation date
from Models.dev: weights availability
from LLMGateway: description
Advantages and trade-offs
Compared with 10 representative transcription models
Advantages
Trade-offs
Costs more than most peers
Maker's price: $0.017 /min, pricier than 5 of 5 priced peers.
No open weights
The weights aren't public, so you can't run it yourself.
Input & output
as stated by the model's providersText
Audio
Text
Also known as: azure/gpt-live-transcribe, openai/gpt-live-transcribe
Like this one. Tap or hover any dotted value, label or icon to see where it comes from and how it is measured. open details when tapped or hovered. How this data works