Giggy Speech to Text

Giggy Speech to Text provides hosted transcription APIs for recorded and streaming audio. Developers can use Batch, Fast and Streaming options, with current pricing and request limits published at https://giggy.ai/speech-to-text. API documentation is available at https://giggy.ai/docs/speech-api.

Giggy also ranks 3rd out of 21 models benchmarked by word error rate in Verli’s independent STT benchmark: https://verli.app/blog/speech-to-text-accuracy-benchmark. This result is specific to the LibriSpeech test-clean evaluation.