whisper_large_v3_turbo

Creates, updates, deletes, gets or lists a whisper_large_v3_turbo resource.

Overview

Name	whisper_large_v3_turbo
Type	Resource
Id	cloudflare.ai.whisper_large_v3_turbo

Fields

The following fields are returned by SELECT queries:

SELECT not supported for this resource, use SHOW METHODS to view available operations for the resource.

Methods

The following methods are available for this resource:

Name	Accessible by	Required Params	Optional Params	Description
workers_ai_post_run_cf_openai_whisper_large_v3_turbo	insert	account_id, audio	queueRequest, tags	Runs inference on the @cf/openai/whisper-large-v3-turbo model.

Parameters

Parameters can be passed in the WHERE clause of a query. Check the Methods section to see which parameters are required or optional for each operation.

Name	Datatype	Description
account_id	string	The Cloudflare account ID.
queueRequest	string
tags	string

INSERT examples

workers_ai_post_run_cf_openai_whisper_large_v3_turbo

Runs inference on the @cf/openai/whisper-large-v3-turbo model.

INSERT INTO cloudflare.ai.whisper_large_v3_turbo (
audio,
beam_size,
compression_ratio_threshold,
condition_on_previous_text,
hallucination_silence_threshold,
initial_prompt,
language,
log_prob_threshold,
no_speech_threshold,
prefix,
task,
vad_filter,
account_id,
queueRequest,
tags
)
SELECT 
'{{ audio }}' /* required */,
{{ beam_size }},
{{ compression_ratio_threshold }},
{{ condition_on_previous_text }},
{{ hallucination_silence_threshold }},
'{{ initial_prompt }}',
'{{ language }}',
{{ log_prob_threshold }},
{{ no_speech_threshold }},
'{{ prefix }}',
'{{ task }}',
{{ vad_filter }},
'{{ account_id }}',
'{{ queueRequest }}',
'{{ tags }}'
;

Manifest

# Description fields are for documentation purposes
- name: whisper_large_v3_turbo
props:
  - name: account_id
    value: "{{ account_id }}"
    description: Required parameter for the whisper_large_v3_turbo resource.
  - name: audio
    value: "{{ audio }}"
    description: |
      Base64 encoded value of the audio data.
  - name: beam_size
    value: {{ beam_size }}
    description: |
      The number of beams to use in beam search decoding. Higher values may improve accuracy at the cost of speed.
    default: 5
  - name: compression_ratio_threshold
    value: {{ compression_ratio_threshold }}
    description: |
      Threshold for filtering out segments with high compression ratio, which often indicate repetitive or hallucinated text.
    default: 2.4
  - name: condition_on_previous_text
    value: {{ condition_on_previous_text }}
    description: |
      Whether to condition on previous text during transcription. Setting to false may help prevent hallucination loops.
    default: true
  - name: hallucination_silence_threshold
    value: {{ hallucination_silence_threshold }}
    description: |
      Optional threshold (in seconds) to skip silent periods that may cause hallucinations.
  - name: initial_prompt
    value: "{{ initial_prompt }}"
    description: |
      A text prompt to help provide context to the model on the contents of the audio.
  - name: language
    value: "{{ language }}"
    description: |
      The language of the audio being transcribed or translated.
  - name: log_prob_threshold
    value: {{ log_prob_threshold }}
    description: |
      Threshold for filtering out segments with low average log probability, indicating low confidence.
    default: -1
  - name: no_speech_threshold
    value: {{ no_speech_threshold }}
    description: |
      Threshold for detecting no-speech segments. Segments with no-speech probability above this value are skipped.
    default: 0.6
  - name: prefix
    value: "{{ prefix }}"
    description: |
      The prefix appended to the beginning of the output of the transcription and can guide the transcription result.
  - name: task
    value: "{{ task }}"
    description: |
      Supported tasks are 'translate' or 'transcribe'.
    default: transcribe
  - name: vad_filter
    value: {{ vad_filter }}
    description: |
      Preprocess the audio with a voice activity detection model.
    default: false
  - name: queueRequest
    value: "{{ queueRequest }}"
  - name: tags
    value: "{{ tags }}"