An audio interview is useful, but hard to search. Azure Speech can turn speech into text, written content into audio, and spoken language into translated captions or dubbing. These capabilities share one Speech services resource. section 7.6
Choose the capability that matches the job:
- Speech to Text transcribes author interviews into searchable archives, making topics and quotes easier to find and reuse. section 7.2
- Text to Speech can generate an immediate audio preview or produce whole chapters in batches. Store reusable files in Azure Blob Storage instead of generating the same audio repeatedly. section 7.2.2
- Speech Translation can provide live multilingual captions or translated audio for recordings. Review translated output before publishing it widely. section 7.2.3
C:/> az cognitiveservices account create --name banningBooksSpeech --resource-group BanningRG --kind SpeechServices --sku S0 --location westus --yes
This Azure CLI command creates a SpeechServices resource in an existing resource group. It selects the S0 pricing tier and West US region; --yes skips confirmation. section 7.2
After provisioning, use the Speech SDKs or APIs for synthesis and translation. Pick a voice and use text or SSML for spoken output; use streaming for live translation and batch processing for recordings. For production, secure credentials and stored results, cache generated audio where practical, and obtain explicit consent before publishing a custom voice resembling a real person. section 7.2.2 section 7.2.3
