Skip to content
Trending storyDeveloping

AssemblyAI launches AssemblyAI-TTS-1 text-to-speech model at $15 per 1M characters

2 articles2 sourcessince Oct 28Last article 52m ago ·

Overview

AISummary of 2 articles

AssemblyAI has launched AssemblyAI-TTS-1, its first text-to-speech model, priced at $15 per 1M characters.

Artificial Analysis ranks it #12 in its Provider Voice Arena with an Elo of 1,227 across 1,626 appearances, #10 on Controlled Voice at 1,110, and #16 on Pronunciation Robustness at 81.7%. It generates 162 characters per second.

LiveKit adds assemblyai-tts-1 to LiveKit Inference, which AssemblyAI's model supports English, Spanish, French, German, Italian, and Portuguese, with more languages planned. LiveKit says the model streams first audio in about 90 ms and offers 4,000+ voices.

Written by AI from the articles below · updated Oct 28, 2:31 PM ET

Check the sources:

Latest developmentOct 28, 2:10 PM ET
LiveKit Inference adds AssemblyAI's assemblyai-tts-1 text-to-speech model

Developments

2 developments

  1. Oct 28, 2:10 PM ET · 1 article
    LiveKit adds AssemblyAI's TTS model assemblyai-tts-1 to LiveKit Inference
    LiveKit: LiveKit Inference adds AssemblyAI's assemblyai-tts-1 text-to-speech model
  2. Oct 28, 12:54 PM ET · 1 article
    AssemblyAI launches AssemblyAI-TTS-1, ranked in Artificial Analysis TTS arenas
    Artificial Analysis: AssemblyAI launches AssemblyAI-TTS-1, ranking #12 on Provider Voice Arena

Article timeline

The articles in this story. Times are ET.

Oct 28
  1. LiveKitOfficialPick
    LiveKit Inference adds AssemblyAI's assemblyai-tts-1 text-to-speech model

    AILiveKit Inference now offers assemblyai-tts-1, the first text-to-speech model from AssemblyAI. The model streams first audio in about 90 ms, offers 4,000+ voices, and supports English, Spanish, French, German, Italian, and Portuguese, costing $15 per 1M characters.

  2. Artificial AnalysisOfficial
    AssemblyAI launches AssemblyAI-TTS-1, ranking #12 on Provider Voice Arena

    AIAssemblyAI has launched AssemblyAI-TTS-1, its first text-to-speech model, which ranks #12 in Artificial Analysis's Provider Voice Arena with an Elo of 1,227 across 1,626 appearances. It ranks #10 on Controlled Voice with a score of 1,110 and #16 on Pronunciation Robustness at 81.7%. The model generates 162 characters per second and costs $15 per 1M characters.

    GIF from @ArtificialAnlys's post

Heat Trend

Not enough continuous observations to show a trend yet.