pyflink.multimodal.expression.AudioExpressionAccessor.detect_language

AudioExpressionAccessor.detect_language(*, top_k: int = 1, model: str = 'openai/whisper-tiny', model_sharing: Optional[Literal['process', 'shared']] = None, concurrency: Optional[int] = None, batch_size: Optional[int] = None, num_gpus: Optional[float] = None, gpu_type: Optional[str] = None) → pyflink.table.expression.Expression

Detect likely spoken language candidates for waveform values.

Equivalent to audio_detect_language(). See that function for full parameter details.

Parameters

• top_k – Number of language candidates to return per row.

• model – Whisper or Whisper-compatible model id.

• model_sharing – Optional model sharing mode.

• concurrency – Optional execution concurrency for this operation. None uses the framework default.

• batch_size – Optional inference batch size.

• num_gpus – Optional number of GPUs requested for model inference.

• gpu_type – GPU type requested for model inference. The DataFrame UDF runtime requires this when num_gpus is set.

Returns

A DataFrame expression that produces ARRAY<ROW<language_tag, language_name, confidence>> language candidates.