> For the complete documentation index, see [llms.txt](https://www.mlcompendium.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://www.mlcompendium.com/generative-ai/speech.md).

# Speech

Generating from speech starts with turning audio into text, and a speech model is only useful once you can measure how often it gets the words wrong. The page goes from the model, Whisper, to the evaluation method, word error rate.

The same notes are in [Basics](https://www.mlcompendium.com/predictive-ml/audio-basics.md), [Deep Neural Audio](https://www.mlcompendium.com/deep-learning/deep-neural-audio.md), [Mix N Match](https://www.mlcompendium.com/generative-ai/mix-n-match.md), and [Other Tools](https://www.mlcompendium.com/predictive-ml/audio-algorithms.md#other-tools).

## Models

The model to start from is OpenAI's Whisper. [Whisper](https://openai.com/research/whisper) is the "Introducing Whisper" announcement, and the code is on [GitHub](https://github.com/openai/whisper) as openai/whisper, "Robust Speech Recognition via Large-Scale Weak Supervision".

## Evaluation Methods

A transcript from that model still has to be scored against a reference.

1. [Jiwer](https://github.com/jitsi/jiwer) - Evaluate your speech-to-text system with similarity measures such as word error rate (WER)


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://www.mlcompendium.com/generative-ai/speech.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
