AvailableOpenAI-compatible transcription

Speech to Text

Turn audio into structured, speaker-aware text.

Submit an uploaded file or a remote URL through one familiar multipart interface. TaskInfer validates inputs up to 100 MB and returns the response format your workflow requests.

Focused input. Predictable output.
InputAudio
OutputTranscript
Speaker diarization

Workflow

A direct path from request to result.

Each product keeps the interface focused on one concrete audio task.

01

Send audio

Use a local upload or place a public HTTPS URL in the same file field.

02

Choose structure

Request text, JSON, timestamps, or speaker-labelled verbose JSON.

03

Use the transcript

Feed consistent output into search, analysis, summaries, or agents.

Capabilities

Built around Speech to Text.

Transcribe multi-speaker audio with timestamps and structured output.

Speaker diarization

Label speakers in verbose JSON for calls, interviews, and meetings.

Precise timestamps

Request segment or word timestamps when downstream timing matters.

Flexible responses

Choose text, JSON, or verbose JSON without changing endpoints.

API

A contract designed for developers.

A multipart transcription request compatible with the familiar OpenAI Whisper request shape.

RequestPOST /v1/audio/transcriptions
curl https://taskinfer.com/v1/audio/transcriptions \
+  -H "Authorization: Bearer $TASKINFER_API_KEY" \
+  -F file=https://cdn.example.com/interview.mp3 \
+  -F response_format=verbose_json

Usage-based pricing

50 credits per started audio second

Each started audio second is billed individually.

View all pricing

Go deeper

Read the full request and response contract.

Open Speech to Text documentation