Speech Services pricing

• For Speech to Text and Speech Translation, usage is billed in one-second increments.
• For Text to Speech: usage is billed per character. Check the definition of character in the pricing note.
• For Speech to Text and Text to Speech, endpoint hosting for custom models is billed per second per model.
• For Custom Commands: billing is tracked as consumption of Speech to Text, Text to Speech, and Language Understanding. Custom Commands does not introduce new billing meters.
• There is no charge for training Speech to Text models. The only costs are endpoint hosting per model once deployed, and then the cost per audio hour of Custom Speech to Text.

The Speech service enables users to adapt baseline models based on their own acoustic and language data, leading to custom speech models that can be used against both Speech to Text and Speech Translation.

The language model is a probability distribution over sequences of words. The language model helps the system decide among sequences of words that sound similar, based on the likelihood of the word sequences themselves. For example, “recognize speech” and “wreck a nice beach” sound alike but the first hypothesis is far more likely to occur, and therefore will be assigned a higher score by the language model. If you expect voice queries to your application to contain particular vocabulary items, such as product names or jargon that rarely occur in typical speech, it is likely that you can obtain improved performance by customizing the language model. For example, if you were building an app to search MSDN by voice, it’s likely that terms like “object-oriented” or “namespace” or “dot net” will appear more frequently than in typical voice applications. Customizing the language model will enable the system to learn this.

The acoustic model is a classifier that labels short fragments of audio into one of several phonemes, or sound units, in each language. These phonemes can then be stitched together to form words. For example, the word “speech” is comprised of four phonemes “s p iy ch”. These classifications are made on the order of 100 times per second. Customizing the acoustic model can enable the system to learn to do a better job recognizing speech in atypical environments. For example, if you have an app designed to be used by workers in a warehouse or factory, a customized acoustic model can more accurately recognize speech in the presence of the noises found in these environments.

Speech service offers a wide range of text-to-speech (TTS) voice fonts, however custom neural voice allows you to build your own custom voice that suits your needs and your brand. Read the blog for more information.

Language identification allows you to identify a switch in spoken language and transcribe speech accordingly. This can be applied in scenarios where the audio language is unknown, or when speaker(s) may speak multiple languages. Single Language Identification is available at no additional cost. Continuous Language Identification is an enhanced add-on feature. Visit docs to learn more.