AI Batch Transcription Benchmark: Transcribing 1 Million+ Hours of Videos in just 7 days for $1800
In this AI batch transcription benchmark, we transcribed 1 Million hrs of YouTube videos for $1800 on SaladCloud (with Parakeet TDT 1.1B).
AI Transcription Benchmark: 1 Million Hours of Youtube Videos with Parakeet TDT 1.1B for Just $1260, a 1000-fold cost reduction
In this benchmark, we transcribe 1 Million hours of YouTube videos in a batch processing mode on SaladCloud for just $1260, a 1000-fold cost reduction.
Text-to-Speech (TTS) API Alternative: Self-Managed OpenVoice vs MetaVoice Comparison
We compare OpenVoice and MetaVoice, showing how self-managed open-source models can deliver ~90% cost savings over text to speech API.
MetaVoice AI Text-to-Speech (TTS) Benchmark: Narrate 100,000 words for only $4.29 on SaladCloud
In this MetaVoice-1B Text-to-Speech benchmark, we compare cost and speed for converting 100,000 words to speech for different GPUs on Salad.
Parakeet TDT 1.1B Inference Benchmark on SaladCloud: 1,000,000 hours of transcription for Just $1260
Inference Benchmark on SaladCloud: Distil-Whisper Large V2 vs. Whisper Large V3 for Speech-to-text
In this speech-to-text inference benchmark, we compare Distil-Whisper Large V2 and Whisper-Large V3 across different clouds and managed APIs.
OpenVoice Text-to-Speech (TTS) Benchmark: 6 Million+ Words/$ Using SaladCloud
In this GPU benchmark for OpenVoice on SaladCloud, consumer GPUs deliver 6 Million+ words per dollar for Text-to-Speech inference.
Whisper Large V3 Speech Recognition Benchmark: 1 Million hours of audio transcription for just $5110
In this Whisper Large v3 benchmark, we show how AI companies can deliver transcription at a 99% lower cost compared to managed services.
Bark Benchmark: Reading 144K Recipes with Text-to-Speech on SaladCloud
In this benchmark of suno-ai/bark, we read 144K recipes with bark's text-to-speech model on Salad's consumer GPUs.
Whisper Large Inference Benchmark: 137 Days of Audio Transcribed in 15 Hours for Just $117
In this inference benchmark for whisper-large-v2 model, we transcribed 137 days of audio in 15 hours for just $117 on SaladCloud.