v1.5.0

⚠️ Breaking Change

The following breaking change to the beta endpoint is introduced in this release:

GoogleDataCollectionConfig and google_data_collection_opt_in have been removed from v1p1beta1.
If your code does not use beta functionality (i.e. you never explicitly require v1p1beta1), your code will not require any changes.

v1.4.0

Features

In this minor release we have added a new optional field enable_automatic_punctuation to RecognitionConfig message in the v1p1beta1 proto. The existing code does not need to be changed, and v1 endpoint is still used by default.

v1.3.0

Features

In this minor release we updated some protos used by v1p1beta1 endpoint (added optional metadata field to RecognitionConfig message). There is no need to change any existing code. Also, v1 endpoint is still default. If you want to try using the new v1p1beta1 endpoint, you need to request it in your code:

const speech = require('@google-cloud/nodejs-speech').v1p1beta1;

Fixes

• (#47) Remove projectId argument in samples

v1.2.0

Fixes

Updated dependencies, including google-gax, to fix streaming behavior. (googleapis/gax-nodejs#197)

v1.1.0

Features

This release introduces a new endpoint v1p1beta1 that introduces some enhancements to speech models. The default endpoint does not change, the existing code will still use v1 endpoint. This change does not affect any existing code.

Use the following code to request v1p1beta1 endpoint for your client:

const speech = require('@google-cloud/speech');
var client = new speech.v1p1beta1.SpeechClient();

v1.0.1

Changes

In this patch release timeouts for Recognize and StreamingRecognize methods were increased.

v1.0.0

Features

General Availability

The cloud speech API client library is now available at the general availability quality level. This means it is stable; the code and API surface will not change in backwards-incompatible ways unless absolutely necessary (e.g. because of critical security issues) or with an extensive deprecation period.

Refer to the nodejs-speech API reference documentation for details.

@google-cloud/speech v0.11.0

This is the (hopefully) final release candidate before declaring a stable release.

⚠️ Breaking Changes!

New client class

The client object is no longer initialized with require('@google-cloud/speech').v1(). This used to return a function that built and returned an internal class.

We now expose an ES6 class directly; there are no wrappers or obfuscation. Basic usage looks like this:

// Import the language module.
const speech = require('@google-cloud/speech');

// Instantiate a client class.
let client = new speech.SpeechClient();

Features

Runnable samples

There are now runnable samples in the samples/ directory. These samples demonstrate the use of this API and offer an additional resource for getting started. These also power the readme, and are automatically tested to ensure that they are up-to-date and correct.

Implementation Details

• Change to a new linter (ESLint) and code style formatter (prettify).
• Update docs to use JSDoc 3.
• Migrate to the googleapis/nodejs-speech repository.

Refer to the nodejs-speech API reference documentation for details.

@google-cloud/speech v0.10.3

Dependencies

• Update google-proto-files (#2592)

@google-cloud/speech v0.10.2

Features

• Add support for WordTimeOffset (#2493)

Dependencies

• Remove google-proto-files dependency. (#2458)