parakeet-ultra is not named as a product on any source page. The closest match is NVIDIA's Parakeet family of speech recognition models, which appears in Replicate's catalogue as dmtanner/parakeet-tdt-0.6b-v3, an ASR model created by Nvidia with word-level timestamps 6. For speech-to-text work generally, Replicate, Hugging Face, ElevenLabs and LocalAI each offer transcription actions or APIs 68910.

1. 1

   Replicatereplicate.comverified and signed

   Runs hosted open-source models via a cloud API, including dmtanner/parakeet-tdt-0.6b-v3, a Nvidia ASR model with word-level timestamps 6.

   WebsiteDocsPricingGet started

   +14 community tools built for it

   for agents: nateraw/whisper-large-v3 · dmtanner/parakeet-tdt-0.6b-v3 · ibm-granite/granite-speech-3.3-8b

2. 2

   Hugging Facehuggingface.coverified and signed

   Hosts open models and datasets and provides an MCP client and Hub SDK for accessing them, including speech models 8.

   WebsiteDocsPricingGet started

   official:MCP server · @huggingface/mcp-clientSDK · huggingface-hubCLI · text-embeddings-inferencerepository · pwc-cliskill · hf-cli+23 community tools built for it

   for agents: models · get_model · train-sentence-transformers

3. 3

   ElevenLabs APIelevenlabs.iofound in code, not yet checked

   Provides speech_to_text and speech_to_speech actions through an API and SDKs for transcribing audio 9.

   WebsiteDocsPricingGet started

   official:MCP server · ElevenLabsAPI · ElevenLabs API Documentation+59 community tools built for it

   for agents: speech_to_text · speech_to_speech · speech-engine

4. 4

   LocalAIlocalai.ioverified and signed

   Open source engine that runs speech models locally, with a transcribe_audio action for turning audio files into text 10.

   WebsiteDocs

   +6 community tools built for it

   for agents: transcribe_audio · text_to_speech · list_models