Benchmarking STT service TTFB and semantic WER for real-time AI applications

Local-first speech AI benchmarking — compare STT, TTS, emotion & diarization models side by side

a place to discuss and create an independent speech-to-text benchmarking tool

No description available

Voice AI benchmarks for all of the latest TTS and STT models.

Code to benchmark your own models against all the competitors on the STT market

⚡ Benchmark LLM, STT, and TTS inference latency across providers. Measure TTFT P50/P95/P99, full voice pipeline benchmarks, CI-ready.

A comprehensive benchmark of Open AI's Whipser STT model in Egyptian ARZ translation to english

Citizen Voice: Speech-to-Text for Grievances -Improve open-source STT for citizen grievance calls (Hindi, English, Hinglish). Benchmark accuracy, implement enhancements (noise reduction, language models).

Speech Streaming STT Benchmarking with Google STT integration

No description available

Benchmark LLM STT

Speech-To-Text (STT)/Transcription Services Benchmarking Tool

Benchmarking bunch of STT both online and offline.

Benchmark demo, STT and TTS testing

Benchmark of common offline stt solutions

cli for benchmarking stt APIs like whisper.cpp, openai etc

R&D benchmarking of Whisper vs Faster-Whisper for STT performance

NextJS React benchmarking app using github.com/pietrop/stt-align-node to compare multiple STT results side by side

Single shot STT benchmark for long from audio

Local CLI for benchmarking STT → TTS pipelines using Fast-Whisper and Piper

Local UI for benchmarking FluidAudio and sherpa STT paths on the same Mac

STT(Speech-To-Text) benchmark and development on Raspberry Pi 4

A cross-platform benchmarking tool for evaluating Speech-to-Text (STT) and Text-to-Speech (TTS) systems.

Benchmark of various sLLMs for post-processing STT transcripts to correct proper noun errors.

MCP server for voice AI development — recommend, benchmark, and scaffold STT+LLM+TTS stacks

Speech pipeline benchmark for NS, VAD, and STT models with unified multi-environment execution, transcript export, and WER/timing evaluation.

Low-latency, local Text-to-Speech integration for LiveKit Agents using Piper ONNX voices. Supports streaming playback, configurable voice/model paths, and realtime latency logging (STT, LLM, TTS) for voice assistant benchmarking.

Citizen Voice: Speech-to-Text for Grievances -Improve open-source STT for citizen grievance calls (Hindi, English, Hinglish). Benchmark accuracy, implement enhancements (noise reduction, language models).

Project for evaluating and improving LLMs, featuring a tool for benchmarking models and a chat system with text + voice support. Deployed on PAL Robotics’ ARI robot with STT, TTS, and real-time audio, enabling natural human-like conversations.