Nova 3
Deepgram Nova-3 is a fast, accurate speech-to-text model served through Telnyx's OpenAI-compatible transcription endpoint. It supports English variants (en, en-US, en-GB, en-AU, en-NZ, en-IN) and only accepts mp3 or wav audio files. Deepgram-specific options such as smart formatting, punctuation, diarization, utterance segmentation, and numeral conversion are configurable via model_config, and word-level timestamps with speaker labels are returned when enabled.
Context
4K
Max output
4K
Input /1M
$0.0001233/s
Output /1M
—
AcceptsAudioProducesTextInstruct type none
Providers
FastRouter routes your requests to this provider. You can also bring your own key.
ProviderInput /1MOutput /1MContextMax outputFeaturesDetails
TE
Telnyx
telnyxInput /1M$0.0001233/s
Output /1M—
Context4K
Max output4K
Supports structured (JSON) output
Supported parameters
Request parameters you can send with this model.
Parameter
Type
Telnyx
Output & Format (Response Formatting)
response_formatStructured output format configuration.
string
"json", "verbose_json"default: json
Other
file
string
-
file_url
string
-
language
string
"en", "en-US", "en-GB", "en-AU", "en-NZ", "en-IN"
timestamp_granularities
string
"segment"
Make your first API call
OpenAI-compatible. Point your SDK at FastRouter, or call the API directly.
Replace
<FASTROUTER_API_KEY> with your key · Get your key →Frequently asked questions
Model scores sourced from ArtificialAnalysis