Here we take an input sample of Kikuyu phrases (and their expert transcription and English translations) and compare the top Speech Reco and Translation models. As of April 3, 2024, we've found that Meta's MMS-Large works best for Kikuyu with Ghana NLP's Kikuyu to English translation services.
. Which AI model actually works best for your needs?Upload your own data and evaluate any Gooey.AI workflow, LLM or AI model against any other.Great for large data sets, AI model evaluation, task automation, parallel processing and automated testing.To get started, paste in a Gooey.AI workflow, upload a CSV of your test data (with header names!), check the mapping of headers to workflow inputs and tap Submit.More tips in the Details below.
Run
Examples
API
Provide one or more Gooey.AI workflow runs.You can add multiple runs from the same recipe (e.g. two versions of your copilot) and we'll run the inputs over both of them.
Speech Recognition and Translation
Whisper Large v3 - Kikuyu
Azure - Kikuyu
Google Chirp - Kikuyu
MMS - Kikuyu
➕ Add a Workflow
Upload or link to a CSV or google sheet that contains your sample input data.For example, for Copilot, this would sample questions or for Art QR Code, would would be pairs of image descriptions and URLs.Remember to includes header names in your CSV too.
Loading...
Submit Links in Bulk
Please select which CSV column corresponds to your workflow's input fields.For the outputs, select the fields that should be included in the output CSV.To understand what each field represents, check out our API docs.
Documents
AUDIO LINKS
Raw Output Text
Run URL
🤲 Show All Columns
Selected Model
———
Language
Translation Model
Source Translation Language
Target Translation Language
Google Translate Target
Glossary Document
Output Format
Price
Run Time
Error Msg
Output Text
(optional) Add one or more Gooey.AI Evaluator Workflows to evaluate the results of your runs.
Low Resource ASR Evaluator
➕ Add an Eval
⚙️ Settings
Run cost = 1 credits
🏃 Submit
By submitting, you agree to Gooey.AI's terms & privacy policy.
https://storage.googleapis.com/dara-c1b52.appspot.com/daras_ai/media/0d9aaee2-f243-11ee-9977-02420a000194/evaluator-19.csv
https://gooey.ai/eval/?run_id=kk18drjggklj&uid=kKZgp2h1H2YxZYxZ2DbiRfUfeDM2
Generated in 767.8s on
...
ℹ️ Details
Building complex AI workflows like copilot) and then evaluating each iteration is complex.Workflows are affected by the particular LLM used (GPT4 vs PalM2), their vector DB knowledge sets (e.g. your google docs), how synthetic data creation happened (e.g. how you transformed your video transcript or PDF into structured data), which translation or speech engine you used and your LLM prompts. Every change can affect the quality of your outputs.
To get started:
🙋🏽♀️ Need more help? Join our Discord