new·The score now tells you which way it movedA brain's exam only ever grows: its own material writes questions, and so does every question a real caller asked and did not get answered. The score is a percentage over that growing set, so a brain that learned more could post a smaller number — and this week three did. One of them answered two MORE questions than the week before and showed eighteen points less. Printed as a single percentage, that reads as decline to a reader and as punishment to anyone who contributes material.all news →
mozg.beta
Sign in

AI SDK · Providers · all subjects

groq/transcription

4 notes, read out of this brain and free to use. Each one was extracted from a source and is re-checked against its exam.

Groq transcription model instantiation

Create a Groq transcription model using the `.transcription()` factory method with the model ID as the first argument: `const model = groq.transcription('whisper-large-v3');`

Groq transcription model options

Groq transcription model provider options: timestampGranularities (string[], defaults to `['segment']`, possible values: `['word']`, `['segment']`, or `['word', 'segment']`, requires `responseFormat` to be `'verbose_json'`); responseFormat (string, set to `'verbose_json'` for timestamps and `timestampGranularities`, or `'text'` for transcribed text only); language (string, ISO-639-1 format like 'en' for improved accuracy and latency); prompt (string, optional text to guide model style or continue previous segment); temperature (number, 0-1, defaults to 0, higher values increase randomness).

Groq transcription example with language option

Example of Groq transcription with language option: `const result = await transcribe({ model: groq.transcription('whisper-large-v3'), audio: await readFile('audio.mp3'), providerOptions: { groq: { language: 'en' } satisfies GroqTranscriptionModelOptions } });`

Groq transcription model capabilities

Groq transcription model capabilities: whisper-large-v3 (Transcription: Yes, Duration: Yes, Segments: Yes, Language: Yes), whisper-large-v3-turbo (Transcription: Yes, Duration: Yes, Segments: Yes, Language: Yes).

Give your agent this brain