Skip to navigation

Limitations

View as Markdown

Read this page before you build on it.

What Dictate does not do

Not offeredWhat to do instead
Subtitle files (SRT, VTT)Build them from section times in your own code, or use Speech-to-Text
Browser tokens. There are no short-lived keys. A browser WebSocket cannot send the key headerOpen the socket from your server
A translate modeAsk for the translation in style.rules
Memory between calls or turns. There is no field for earlier turnsAdd names to vocabulary as they are said (config.update on the socket)
A recording over 1 hourSplit the file, and send each part as a job

Limits

LimitValue
Clip length and size30 s, up to 10,000,000 bytes
Job length1 hour. An upload under about 100 MB. Use audio_url for more
Job results are kept7 days. Delete a job when you have the result
Live session3 hours. Closes after 60 s with no message from you. Each message up to 1 MiB, and at most 1 s of audio (0.5 s with stream_type=fast)
vocabularyUp to 1,000 entries
context.reference / context.text50,000 / 2,000 characters
style.rules2,000 characters
output_rules4,000 characters
Other form fieldsAt most 1 MiB each
Rate limitsPer account, the same as Speech-to-Text. See the pricing page