Speech-to-text, TTS, and audio analysis
10
Models
3
Benchmarks
6
Providers
7
Open Weight
9
Deployable Now
7 self-host
10
Structured benchmark-backed rows
0
Signal-backed rows without full normalized tables
0
Benchmark-expected rows still catching up
| # | Model | Adoption | Quality | Params | GAIA | SWE-Bench | Aider Polyglot | $/M In | Speed | Open |
|---|---|---|---|---|---|---|---|---|---|---|
6 | qwen3-tts Qwen Open Weights |
| 67.2 |
| 46.1 |
| Unknown |
| 44.2 |
| 69.6 |
| — |
Free Free |
| — |
16 | Qwen3-ASR-1.7B Qwen Open WeightsStructured | 64.5 | 51.0 | 2B | 44.2 | 69.6 | 8.0 | Free Free | — |
17 | Qwen3-TTS-12Hz-1.7B-CustomVoice Qwen Open WeightsStructured | 64.1 | 50.5 | 2B | 44.2 | 69.6 | 8.0 | Free Free | — |
2 | GPT Audio OpenAI SubscriptionStructured | 63.0 | 17.6 | Undisclosed | — | 42.4 | — | $2.50/M Cheapest verified | — |
18 | TTS-1 HD OpenAI Structured | 63.0 | 13.4 | Undisclosed | — | 42.4 | — | $30.00/M Cheapest verified | — |
9 | GPT-4o Transcribe Diarize OpenAI SubscriptionStructured | 60.2 | 19.5 | Undisclosed | 60.6 | — | — | $2.50/M Cheapest verified | — |
19 | cohere-transcribe-03-2026 Cohere Open WeightsStructured | 58.9 | 47.7 | Unknown | — | 42.4 | — | Free Free | — |
22 | Voxtral-4B-TTS-2603 Mistral AI Open WeightsStructured | 54.7 | 41.1 | 4B | — | 42.4 | — | Free Free | — |
14 | qwenasr twangodev Open WeightsStructuredArchived | — | — | Unknown | — | 69.6 | — | Free Free | — |
10 | granite-4.0-1b-speech ibm-granite Open WeightsStructured | — | — | 1B | — | 42.4 | — | Free Free | — |