click any column header to sort
← SpeedSpeedSamples → · all runs

TTS Bench — Samples — windows-cloning

Rig: windows-5090 — AMD Ryzen 9 9950X3D 16-Core Processor (16C) · NVIDIA GeForce RTX 5090 32GB · 126 GB RAM · Windows 11
Label: cloning · chris_hemsworth_15s — ref chris_hemsworth_15s.wav
5 prompt(s) · one section per prompt · all models ranked by warm TTFA (fastest first) within each
Each prompt section shows every model's audio output, ordered by warm TTFA (fastest first). Click any audio player to hear that model's rendering.

Reference voice

Each model below was given this clip + transcript as the voice to imitate. Source: chris_hemsworth_15s.wav

Prompt 1

[en]"Open the browser and read my email."
Rank Model Device TTFA warm Audio
1 Pocket-TTS cpu 63ms
2 Sopro V2 Turbo (streaming) cuda 121ms
3 StyleTTS 2 cpu 129ms
4 StyleTTS 2 cuda 140ms
5 OpenVoice v2 cuda 185ms
6 Audio8 TTS 0.6B (compiled) cuda 258ms
7 Sopro V2 Turbo cuda 279ms
8 NeuTTS Nano cuda 356ms
9 NeuTTS Nano cpu 369ms
10 Breeze TTS 2 cuda 426ms
11 NeuTTS Air cuda 497ms
12 Qwen3-TTS 1.7B (CUDA-graph) cuda 529ms
13 Coqui XTTS-v2 cuda 563ms
14 OpenVoice v2 cpu 623ms
15 NeuTTS Air cpu 630ms
16 Chatterbox Turbo cuda 632ms
17 MiraTTS cuda 645ms
18 F5-TTS v1 cuda 728ms
19 OmniVoice cuda 736ms
20 ZipVoice 123M cuda 1.08s
21 Sopro V2 Turbo (streaming) cpu 1.15s
22 Chatterbox cuda 1.25s
23 Dia 1.6B-0626 cuda 1.30s
24 LongCat-AudioDiT 1B cuda 1.44s
25 VibeVoice 7B cuda 1.68s
26 VoxCPM2 2B cuda 1.81s
27 WavTTS 0.67B cuda 1.97s
28 IndexTTS-2 cpu 2.09s
29 Fish Speech 1.5 cuda 2.16s
30 LongCat-AudioDiT 3.5B cuda 2.19s
31 MOSS-TTS-Nano cuda 2.28s
32 Echo-TTS cuda 2.29s
33 Sopro V2 Turbo cpu 2.29s
34 Coqui XTTS-v2 cpu 2.53s
35 IndexTTS-2 cuda 2.57s
36 MOSS-TTS-Nano cpu 2.61s
37 MOSS-TTS v1.5 cuda 2.98s
38 MOSS-TTS v1.0 cuda 2.99s
39 Zonos v0.1 cuda 3.41s
40 Chatterbox Turbo cpu 3.48s
41 Sesame CSM-1B cuda 3.97s
42 Audio8 TTS Preview 0.6B cuda 4.09s
43 DramaBox cuda 4.12s
44 Audio8 TTS Preview 0.1B cuda 5.79s
45 OuteTTS 1.0 1B cuda 6.26s
46 VoxCPM2 2B cpu 6.39s
47 Miso TTS 8B cuda 6.46s
48 Chatterbox cpu 6.52s
49 ZipVoice 123M cpu 7.67s
50 Fish Speech 1.5 cpu 11.56s
51 Audio8 TTS Preview 0.1B cpu 15.91s
52 Audio8 TTS Preview 0.6B cpu 16.50s
53 Sesame CSM-1B cpu 17.17s
54 Zonos v0.1 cpu 18.60s
55 Mars5-TTS cuda 29.79s
56 Mars5-TTS cpu 30.56s
57 OmniVoice cpu 37.89s
58 F5-TTS v1 cpu 50.19s

Prompt 2

[en]"I'll start a new git branch, push the changes, and open a pull request when the tests pass."
Rank Model Device TTFA warm Audio
1 Pocket-TTS cpu 86ms
2 Sopro V2 Turbo (streaming) cuda 118ms
3 StyleTTS 2 cuda 173ms
4 StyleTTS 2 cpu 173ms
5 OpenVoice v2 cuda 339ms
6 Breeze TTS 2 cuda 357ms
7 NeuTTS Nano cuda 368ms
8 Sopro V2 Turbo cuda 407ms
9 NeuTTS Nano cpu 460ms
10 Audio8 TTS 0.6B (compiled) cuda 463ms
11 NeuTTS Air cpu 495ms
12 NeuTTS Air cuda 496ms
13 Sopro V2 Turbo (streaming) cpu 738ms
14 OmniVoice cuda 759ms
15 F5-TTS v1 cuda 914ms
16 MiraTTS cuda 1.08s
17 Coqui XTTS-v2 cuda 1.09s
18 OpenVoice v2 cpu 1.09s
19 Chatterbox Turbo cuda 1.13s
20 Qwen3-TTS 1.7B (CUDA-graph) cuda 1.22s
21 LongCat-AudioDiT 1B cuda 1.28s
22 LongCat-AudioDiT 3.5B cuda 1.93s
23 Echo-TTS cuda 2.16s
24 WavTTS 0.67B cuda 2.19s
25 Chatterbox cuda 2.21s
26 Sopro V2 Turbo cpu 2.93s
27 MOSS-TTS-Nano cuda 3.38s
28 VoxCPM2 2B cuda 3.43s
29 VibeVoice 7B cuda 3.50s
30 MOSS-TTS v1.0 cuda 3.52s
31 MOSS-TTS v1.5 cuda 3.83s
32 MOSS-TTS-Nano cpu 3.97s
33 IndexTTS-2 cpu 4.01s
34 IndexTTS-2 cuda 4.12s
35 DramaBox cuda 4.63s
36 Fish Speech 1.5 cuda 5.49s
37 Coqui XTTS-v2 cpu 6.01s
38 Chatterbox Turbo cpu 6.28s
39 Audio8 TTS Preview 0.6B cuda 7.52s
40 Zonos v0.1 cuda 7.85s
41 Dia 1.6B-0626 cuda 8.34s
42 Sesame CSM-1B cuda 9.01s
43 Chatterbox cpu 11.99s
44 Audio8 TTS Preview 0.1B cuda 12.73s
45 ZipVoice 123M cpu 13.40s
46 OuteTTS 1.0 1B cuda 13.48s
47 VoxCPM2 2B cpu 14.44s
48 Audio8 TTS Preview 0.1B cpu 28.29s
49 Fish Speech 1.5 cpu 30.34s
50 Miso TTS 8B cuda 33.31s
51 Audio8 TTS Preview 0.6B cpu 33.62s
52 Mars5-TTS cuda 33.89s
53 Sesame CSM-1B cpu 38.15s
54 Mars5-TTS cpu 43.83s
55 Zonos v0.1 cpu 47.31s
56 OmniVoice cpu 48.65s
57 F5-TTS v1 cpu 64.77s
58 ZipVoice 123M cuda 133.84s

Prompt 3

[en]"The Parakeet TDT zero point six billion parameter model achieves one point six nine percent word error rate on LibriSpeech test-clean, beating Whisper Large V3 at two point seven percent while running at over two thousand times realtime on a single GPU."
Rank Model Device TTFA warm Audio
1 Pocket-TTS cpu 105ms
2 Sopro V2 Turbo (streaming) cuda 111ms
3 StyleTTS 2 cuda 306ms
4 NeuTTS Nano cuda 307ms
5 StyleTTS 2 cpu 340ms
6 Breeze TTS 2 cuda 376ms
7 NeuTTS Nano cpu 378ms
8 NeuTTS Air cuda 518ms
9 OpenVoice v2 cuda 525ms
10 NeuTTS Air cpu 636ms
11 Sopro V2 Turbo (streaming) cpu 728ms
12 OmniVoice cuda 934ms
13 Audio8 TTS 0.6B (compiled) cuda 1.19s
14 Sopro V2 Turbo cuda 1.32s
15 LongCat-AudioDiT 1B cuda 1.38s
16 F5-TTS v1 cuda 1.71s
17 Echo-TTS cuda 2.14s
18 MiraTTS cuda 2.62s
19 LongCat-AudioDiT 3.5B cuda 2.81s
20 OpenVoice v2 cpu 3.05s
21 Chatterbox Turbo cuda 3.27s
22 Coqui XTTS-v2 cuda 3.35s
23 Qwen3-TTS 1.7B (CUDA-graph) cuda 3.37s
24 WavTTS 0.67B cuda 4.91s
25 Chatterbox cuda 5.20s
26 DramaBox cuda 6.22s
27 MOSS-TTS v1.0 cuda 8.96s
28 IndexTTS-2 cpu 9.05s
29 MOSS-TTS-Nano cuda 9.17s
30 MOSS-TTS v1.5 cuda 9.43s
31 VibeVoice 7B cuda 9.65s
32 Sopro V2 Turbo cpu 10.15s
33 VoxCPM2 2B cuda 10.29s
34 MOSS-TTS-Nano cpu 10.40s
35 IndexTTS-2 cuda 10.75s
36 Sesame CSM-1B cuda 17.44s
37 Chatterbox Turbo cpu 20.05s
38 Fish Speech 1.5 cuda 20.28s
39 Coqui XTTS-v2 cpu 20.33s
40 Audio8 TTS Preview 0.6B cuda 23.47s
41 Zonos v0.1 cuda 23.91s
42 Dia 1.6B-0626 cuda 24.69s
43 Chatterbox cpu 34.74s
44 VoxCPM2 2B cpu 35.11s
45 OuteTTS 1.0 1B cuda 39.66s
46 Audio8 TTS Preview 0.1B cuda 41.75s
47 Mars5-TTS cpu 54.81s
48 Mars5-TTS cuda 55.86s
49 Miso TTS 8B cuda 68.73s
50 Audio8 TTS Preview 0.1B cpu 75.68s
51 Sesame CSM-1B cpu 77.75s
52 OmniVoice cpu 78.95s
53 F5-TTS v1 cpu 103.89s
54 Audio8 TTS Preview 0.6B cpu 104.15s
55 Fish Speech 1.5 cpu 107.15s
56 Zonos v0.1 cpu 157.87s
57 ZipVoice 123M cpu 183.07s

Prompt 4

[en]"Run pytest tests slash test underscore voice dot py with verbose flag and capture flag set to no."
Rank Model Device TTFA warm Audio
1 Pocket-TTS cpu 95ms
2 Sopro V2 Turbo (streaming) cuda 118ms
3 StyleTTS 2 cuda 228ms
4 StyleTTS 2 cpu 230ms
5 NeuTTS Nano cuda 337ms
6 OpenVoice v2 cuda 378ms
7 NeuTTS Nano cpu 386ms
8 Breeze TTS 2 cuda 440ms
9 NeuTTS Air cuda 458ms
10 Sopro V2 Turbo cuda 483ms
11 NeuTTS Air cpu 512ms
12 Audio8 TTS 0.6B (compiled) cuda 694ms
13 Sopro V2 Turbo (streaming) cpu 739ms
14 OmniVoice cuda 752ms
15 F5-TTS v1 cuda 870ms
16 OpenVoice v2 cpu 1.30s
17 Chatterbox Turbo cuda 1.32s
18 LongCat-AudioDiT 1B cuda 1.35s
19 MiraTTS cuda 1.48s
20 Qwen3-TTS 1.7B (CUDA-graph) cuda 1.54s
21 Coqui XTTS-v2 cuda 1.59s
22 WavTTS 0.67B cuda 2.29s
23 Echo-TTS cuda 2.34s
24 LongCat-AudioDiT 3.5B cuda 2.34s
25 Chatterbox cuda 2.87s
26 MOSS-TTS-Nano cuda 3.69s
27 IndexTTS-2 cpu 4.32s
28 MOSS-TTS-Nano cpu 4.58s
29 VibeVoice 7B cuda 4.60s
30 MOSS-TTS v1.0 cuda 4.69s
31 Sopro V2 Turbo cpu 4.71s
32 DramaBox cuda 4.98s
33 IndexTTS-2 cuda 5.19s
34 VoxCPM2 2B cuda 5.19s
35 MOSS-TTS v1.5 cuda 6.51s
36 Fish Speech 1.5 cuda 7.20s
37 Chatterbox Turbo cpu 7.65s
38 Zonos v0.1 cuda 8.48s
39 Coqui XTTS-v2 cpu 9.36s
40 Dia 1.6B-0626 cuda 10.53s
41 Audio8 TTS Preview 0.6B cuda 11.09s
42 VoxCPM2 2B cpu 14.38s
43 Sesame CSM-1B cuda 15.42s
44 Chatterbox cpu 16.32s
45 ZipVoice 123M cpu 16.61s
46 OuteTTS 1.0 1B cuda 17.15s
47 Audio8 TTS Preview 0.1B cuda 18.57s
48 Miso TTS 8B cuda 21.77s
49 Fish Speech 1.5 cpu 36.09s
50 Mars5-TTS cuda 38.59s
51 Audio8 TTS Preview 0.1B cpu 40.80s
52 Mars5-TTS cpu 43.71s
53 Audio8 TTS Preview 0.6B cpu 45.06s
54 Zonos v0.1 cpu 49.51s
55 OmniVoice cpu 49.71s
56 Sesame CSM-1B cpu 61.47s
57 F5-TTS v1 cpu 65.87s
58 ZipVoice 123M cuda 142.27s

Prompt 5

[fr]"Bonjour, je m'appelle Cicero et je vais vous aider avec votre code aujourd'hui."
Rank Model Device TTFA warm Audio
1 Sopro V2 Turbo (streaming) cuda 130ms
2 Pocket-TTS cpu 273ms
3 NeuTTS Nano cuda 289ms
4 Sopro V2 Turbo cuda 295ms
5 NeuTTS Nano cpu 342ms
6 OpenVoice v2 cuda 379ms
7 Audio8 TTS 0.6B (compiled) cuda 446ms
8 Sopro V2 Turbo (streaming) cpu 727ms
9 OmniVoice cuda 833ms
10 Coqui XTTS-v2 cuda 953ms
11 OpenVoice v2 cpu 1.12s
12 Qwen3-TTS 1.7B (CUDA-graph) cuda 1.20s
13 Sopro V2 Turbo cpu 2.22s
14 VoxCPM2 2B cuda 2.87s
15 MOSS-TTS v1.0 cuda 3.52s
16 MOSS-TTS-Nano cpu 3.61s
17 MOSS-TTS v1.5 cuda 3.64s
18 Zonos v0.1 cuda 5.57s
19 Coqui XTTS-v2 cpu 5.58s
20 MOSS-TTS-Nano cuda 5.84s
21 Fish Speech 1.5 cuda 5.93s
22 Audio8 TTS Preview 0.6B cuda 7.10s
23 Audio8 TTS Preview 0.1B cuda 9.07s
24 VoxCPM2 2B cpu 9.90s
25 OuteTTS 1.0 1B cuda 10.67s
26 ZipVoice 123M cpu 14.56s
27 Audio8 TTS Preview 0.1B cpu 27.42s
28 Fish Speech 1.5 cpu 28.17s
29 Audio8 TTS Preview 0.6B cpu 28.67s
30 Zonos v0.1 cpu 30.92s
31 OmniVoice cpu 51.20s
32 ZipVoice 123M cuda 146.78s