> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://onepin.ai/docs/workflow/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://onepin.ai/docs/_mcp/server.

# Workflow

> Onepin is a voice workflow platform that orchestrates, validates, and ships production-ready audio across 51 TTS models.

## Docs

- [Workflow](https://onepin.ai/docs/workflow/overview.md): How an Onepin workflow definition works — the graph of Source, Operator, Validator, and Sink nodes, how a line of script becomes finished audio, and how to edit a definition in code.
- [General rules](https://onepin.ai/docs/workflow/general-rules.md): The rules an Onepin workflow definition must satisfy — required nodes, port compatibility, connectedness, validator routing, and retry loops.
- [Script](https://onepin.ai/docs/workflow/script.md): The Onepin Script node (source_script) — the source of every line. Configure inline text or an uploaded .txt/.pdf file, plus the source language.
- [Translator](https://onepin.ai/docs/workflow/translator.md): The Onepin Translator node (operator_translator) — translate a script into up to 30 target languages, with a Generator downstream for each one.
- [Normalizer](https://onepin.ai/docs/workflow/normalizer.md): The Onepin Normalizer node (operator_normalizer) — expands numbers, dates, and symbols into their spoken form before generation.
- [Generator](https://onepin.ai/docs/workflow/generator.md): The Onepin Generator node (operator_generator) — the TTS engine. Configure voice_map to assign a provider voice and a catalog model per language, and get the two identifiers right.
- [Phoneme Injector](https://onepin.ai/docs/workflow/phoneme-injector.md): The Onepin Phoneme Injector node (operator_phoneme_injector) — names the words worth checking, looks them up in the built-in pronunciation reference, and attaches per-word pronunciation to each line.
- [Pronunciation Corrector](https://onepin.ai/docs/workflow/pronunciation-corrector.md): The Onepin Pronunciation Corrector node (operator_pronunciation_corrector) — replaces only the words the Pronunciation validator flagged, keeping the surrounding audio and the voice unchanged.
- [Validators](https://onepin.ai/docs/workflow/validators.md): How Onepin validators work — the pass and fail outputs, wiring pass to Export and fail back to the Generator, what retry and threshold mean, and chaining vs parallel checks.
- [Accuracy validator](https://onepin.ai/docs/workflow/accuracy-validator.md): The Onepin Accuracy validator (validator_error_rate) — transcribes generated audio and compares it to the script, scoring word accuracy and retrying failing lines.
- [Naturalness validator](https://onepin.ai/docs/workflow/naturalness-validator.md): The Onepin Naturalness validator (validator_naturalness) — scores generated speech for naturalness on a 0–100 scale and retries lines below the threshold.
- [Noise validator](https://onepin.ai/docs/workflow/noise-validator.md): The Onepin Noise validator (validator_noise) — scores generated audio for clarity on a 0–100 scale, flagging background noise and artifacts, and retries lines below the threshold.
- [Pronunciation validator](https://onepin.ai/docs/workflow/pronunciation-validator.md): The Onepin Pronunciation validator (validator_pronunciation) — scores segmental pronunciation phoneme by phoneme, producing a 0–100 score from the phoneme error rate, and retries lines below the threshold.
- [Export](https://onepin.ai/docs/workflow/export.md): The Onepin Export node (sink_preview) — produces the final audio files. Choose wav or mp3; the run's download is a ZIP of the audio plus a manifest.csv.