Unfragile

The request names the model dmtanner/parakeet-tdt-0.6b-v3, an ASR model created by Nvidia that takes .wav inputs or m3u8 URLs with word-level timestamps 6. It is run through Replicate, whose page offers a cloud API for hosted open-source models 1, and Nvidia's own Parakeet TDT speech-to-text offering is listed as an alternative 2.

1
Replicatereplicate.comverified and signed
Runs open-source machine learning models, including dmtanner/parakeet-tdt-0.6b-v3, through a cloud API 16.
Website Docs Pricing Get started
for agents: dmtanner/parakeet-tdt-0.6b-v3
2
NVIDIA Parakeet TDT 0.6B (Brev.dev)nvidia.comverified and signed
Provides Parakeet TDT 0.6B speech-to-text via Nvidia's Brev.dev console for automatic speech recognition 27.
Website Docs Pricing Get started
for agents: automatic-speech-recognition · speech-to-text · nim
3
ElevenLabselevenlabs.iofound in code, not yet checked
Transcribes audio to text with its speech-to-text action for subtitles and meeting transcription 49.
Website Docs Pricing Get started
official:MCP server · ElevenLabs+58 community tools built for it
for agents: speech_to_text · speech-to-text · speech_to_speech
4
LocalAIlocalai.ioverified and signed
Transcribes audio files into text with a transcribe action in an open source engine that runs on your own hardware 510.
Website Docs
+6 community tools built for it
for agents: transcribe_audio · text_to_speech · create_embedding

Sources

1Replicate - Run AI with an APIreplicate.com · read 2026-10-29
2Brev.dev Consolebrev.nvidia.com · read 2026-10-29
4Free AI Voice Generator & Voice Agents Platform | ElevenLabselevenlabs.io · read 2026-10-29
5LocalAI · Make AI run on every machinelocalai.io · read 2026-10-29
6dmtanner/parakeet-tdt-0.6b-v3: declares "dmtanner/parakeet-tdt-0.6b-v3" (the record we read and signed)unfragile.ai · read 2026-10-24
7Parakeet TDT 0.6B v2 Speech to Text: declares "automatic-speech-recognition" (the record we read and signed)unfragile.ai · read 2026-10-24
9ElevenLabs MCP Server: declares "speech_to_text" (the record we read)github.com · read 2026-10-24
10LocalAI (Activepieces): declares "transcribe_audio" (the record we read and signed)unfragile.ai · read 2026-10-24

For agents: the 10 actions behind this answer

dmtanner/parakeet-tdt-0.6b-v3 · dmtanner/parakeet-tdt-0.6b-v3 verified and signed
ASR model, created by Nvidia, with word-level timestamps available. Supports .wav inputs, or m3u8 urls, with a start and end time (to only process a section of the m3u8).
automatic-speech-recognition · Parakeet TDT 0.6B v2 Speech to Text verified and signed
Task: Automatic Speech Recognition
speech-to-text · Parakeet TDT 0.6b En-US verified and signed
Task: Speech to Text
automatic-speech-recognition · Parakeet CTC 0.6B Mandarin English ASR verified and signed
Task: Automatic Speech Recognition
nvidia/parakeet-rnnt-1.1b · nvidia/parakeet-rnnt-1.1b verified and signed
🗣️ Nvidia + Suno.ai's speech-to-text conversion with high accuracy and efficiency 📝
automatic-speech-recognition · Parakeet CTC 0.6B Vietnames English ASR verified and signed
Task: Automatic Speech Recognition
h100-80gb-hbm3 · Parakeet 1.1b RNNT Multilingual Speech to Text verified and signed
Task: H100 80GB HBM3
translate · OpenAI (Activepieces) verified and signed
Translate Audio — Turn speech in any language into English text.
speech_to_text · ElevenLabs MCP Server found in code, not yet checked
transcribe_audio · LocalAI (Activepieces) verified and signed
Transcribe Audio — Turn speech in an audio file into text.

Labels are the instruments' verdict and are never changed by ranking. Agents and robots get this same answer as JSON at /api/v1/find?q=…&answer=1 and through the MCP server.