Rig:mac-m4 — Apple M4 (10C) · Apple M4 GPU (MPS) · 16 GB RAM · Darwin 25.5.0
Label: default voice
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.
Prompt 1
[en]"Open the browser and read my email."
Rank
Model
Device
TTFA warm
Audio
1
Pocket-TTS
cpu
30ms
2
Piper
cpu
62ms
3
Kokoro
mps
193ms
4
NeuTTS Nano
cpu
279ms
5
Kokoro
cpu
299ms
6
Soprano 80M
cpu
324ms
7
KittenTTS
cpu
331ms
8
NeuTTS Air
cpu
361ms
9
Soprano 80M
mps
427ms
10
NeuTTS Nano
mps
485ms
11
NeuTTS Air
mps
576ms
12
Supertonic
cpu
681ms
13
Coqui XTTS-v2
mps
1.15s
14
Coqui XTTS-v2
cpu
1.53s
15
Chatterbox Turbo
mps
1.71s
16
Chatterbox Turbo
cpu
2.47s
17
VibeVoice Realtime 0.5B
mps
2.64s
18
OmniVoice
mps
2.64s
19
OmniVoice
cpu
5.20s
20
Chatterbox
mps
5.94s
21
Chatterbox
cpu
5.97s
22
Magpie-TTS
cpu
6.27s
23
VibeVoice Realtime 0.5B
cpu
6.98s
24
Qwen3-TTS 1.7B
cpu
11.52s
25
Sesame CSM-1B
cpu
18.63s
26
F5-TTS
mps
31.55s
27
VoxCPM2 2B
cpu
34.26s
28
F5-TTS
cpu
34.61s
29
IndexTTS-2
cpu
34.89s
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
31ms
2
Piper
cpu
182ms
3
NeuTTS Nano
cpu
274ms
4
NeuTTS Air
cpu
353ms
5
Kokoro
mps
384ms
6
NeuTTS Nano
mps
485ms
7
NeuTTS Air
mps
562ms
8
Kokoro
cpu
575ms
9
Soprano 80M
cpu
685ms
10
KittenTTS
cpu
726ms
11
Soprano 80M
mps
937ms
12
Supertonic
cpu
1.26s
13
Coqui XTTS-v2
mps
3.09s
14
Chatterbox Turbo
mps
3.74s
15
Coqui XTTS-v2
cpu
3.87s
16
Chatterbox Turbo
cpu
4.27s
17
VibeVoice Realtime 0.5B
mps
5.77s
18
OmniVoice
mps
6.03s
19
OmniVoice
cpu
9.43s
20
Chatterbox
cpu
12.49s
21
Magpie-TTS
cpu
15.34s
22
VibeVoice Realtime 0.5B
cpu
17.14s
23
Sesame CSM-1B
cpu
27.14s
24
Qwen3-TTS 1.7B
cpu
27.96s
25
IndexTTS-2
cpu
33.89s
26
VoxCPM2 2B
cpu
35.93s
27
Chatterbox
mps
36.06s
28
F5-TTS
mps
39.61s
29
F5-TTS
cpu
42.36s
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
37ms
2
NeuTTS Nano
cpu
281ms
3
NeuTTS Air
cpu
366ms
4
Piper
cpu
451ms
5
NeuTTS Nano
mps
519ms
6
NeuTTS Air
mps
611ms
7
Kokoro
mps
1.09s
8
Kokoro
cpu
1.69s
9
Soprano 80M
cpu
1.84s
10
KittenTTS
cpu
2.21s
11
Soprano 80M
mps
2.80s
12
Supertonic
cpu
3.34s
13
Chatterbox Turbo
mps
10.61s
14
Coqui XTTS-v2
cpu
13.18s
15
Chatterbox Turbo
cpu
13.70s
16
VibeVoice Realtime 0.5B
mps
16.69s
17
Chatterbox
mps
19.39s
18
OmniVoice
cpu
23.35s
19
Coqui XTTS-v2
mps
26.13s
20
Chatterbox
cpu
33.99s
21
Sesame CSM-1B
cpu
40.11s
22
Qwen3-TTS 1.7B
cpu
48.35s
23
VibeVoice Realtime 0.5B
cpu
53.07s
24
Magpie-TTS
cpu
67.99s
25
F5-TTS
cpu
68.69s
26
F5-TTS
mps
70.52s
27
IndexTTS-2
cpu
92.13s
28
VoxCPM2 2B
cpu
100.17s
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
34ms
2
Piper
cpu
191ms
3
NeuTTS Nano
cpu
274ms
4
NeuTTS Air
cpu
354ms
5
Kokoro
mps
453ms
6
NeuTTS Nano
mps
478ms
7
NeuTTS Air
mps
584ms
8
Kokoro
cpu
694ms
9
KittenTTS
cpu
769ms
10
Soprano 80M
cpu
771ms
11
Soprano 80M
mps
1.05s
12
Supertonic
cpu
1.45s
13
Chatterbox Turbo
mps
4.21s
14
Chatterbox Turbo
cpu
5.34s
15
Coqui XTTS-v2
cpu
5.78s
16
OmniVoice
mps
6.32s
17
Coqui XTTS-v2
mps
6.73s
18
VibeVoice Realtime 0.5B
mps
8.11s
19
OmniVoice
cpu
9.70s
20
Chatterbox
cpu
15.85s
21
Qwen3-TTS 1.7B
cpu
20.84s
22
VibeVoice Realtime 0.5B
cpu
23.21s
23
Magpie-TTS
cpu
23.35s
24
IndexTTS-2
cpu
32.67s
25
Sesame CSM-1B
cpu
39.31s
26
F5-TTS
mps
40.49s
27
F5-TTS
cpu
42.94s
28
Chatterbox
mps
64.01s
29
VoxCPM2 2B
cpu
69.00s
Prompt 5
[fr]"Bonjour, je m'appelle Cicero et je vais vous aider avec votre code aujourd'hui."