Start with small sets of sample data that match the language, acoustics, and hardware where your model will be used. Training with text is much faster than training with audio (minutes versus days). This data will improve the recognition of special terms and phrases. Start with plain-text data or structured-text data. Training with plain text or structured text usually finishes within a few minutes. Up to 10 classes with up to 4,000 items and up to 50,000 training sentences Data requirements will vary depending on whether you're creating a test or training a model. Not every data type is required to create a model.

The following table lists accepted data types, when each data type should be used, and the recommended quantity. Including data that isn't within your custom model's recognition requirements can harm recognition quality overall. Only include data that your model needs to transcribe.You can add more data to your model later. Keep the dataset diverse and representative of your project requirements.If your model must identify speech recorded on devices of varying quality, the audio data that you provide to train your model must also represent these diverse scenarios. Record audio with hardware devices that the production system will use.Include samples from different environments, for example, indoor, outdoor, and road noise, where your model will be used.Many factors can vary speech, including accents, dialects, language-mixing, age, gender, voice pitch, stress level, and time of day. Include all speech variances that you want your model to recognize.For example, a model that raises and lowers the temperature needs training on statements that people might make to request such changes. Include text and audio data to cover the kinds of verbal statements that your users will make when they're interacting with your model.Consider these factors when you're gathering data for custom model testing and training: Text and audio that you use to test and train a custom model should include samples from a diverse set of speakers and scenarios that you want your model to recognize. This article covers the types of training and testing data that you can use for Custom Speech. In a Custom Speech project, you can upload datasets for training, qualitative inspection, and quantitative measurement.

Azure speech to text custom model