Rendered at 09:24:11 GMT+0000 (Coordinated Universal Time) with Cloudflare Workers.
asaiacai 2 days ago [-]
This is really cool work! I'm curious like what do you see as the biggest lever for speeding up TTS models or from a technical perspective that this was a promising direction in the first place to push on. If I were to guess, some distillation but I'm certain there are probably TTS model aware architectural changes that just make inference wayyyy faster?
awesome! will look into Darwin TTS. super interesting
recentlypostedj 18 hours ago [-]
Question: How do you plan to differentiate, because there are so many TTS and its constantly changing every month who would become better
karimf 1 days ago [-]
This is awesome. Thanks for pushing the audio pareto frontier forward.
Probably far fetched for now, but I think the next big evolution is building the pareto/much cheaper alternative to GPT-Live-1.
The STT/TTS market is quite saturated, while today, there's almost no cheap/open source alternative to GPT-Live-1.
6879346626 1 days ago [-]
Is this really something people want? Honestly you can properly lower the pricing at least by 50%+.
Getting something conversationally better has been done, the tool calling will likely be worse though.
The infrastructure for real time is really annoying though.
toebee 23 hours ago [-]
if we can lower the pricing by not 50% but 10x, then I think it would be something people want. we are taking the bet that OSS models will take a huge chunk of market share not just in LLMs but in multimodal as well
6879346626 15 hours ago [-]
If audio only then 10x is 100% possible right now based on math of the services.
+ Video is unlikely unless they are willing to give up margins.
Didn't take long at all for Google to smash out with 50% lol
toebee 1 days ago [-]
agreed. we've been doing some work around NVIDIA personaplex 7b, but its quality is quite far from GPT-Live-1, esp in terms of intelligence. Once a good OSS model is out, we'll be sure to be the first to serve it cheaply to the masses :)
rahimnathwani 2 days ago [-]
For some reason it switched voices half way through a 33 second clip.
For OP the clip name is nari-nina-01a0a12f-980a-765e-8029-fa56bd23210d.wav
toebee 1 days ago [-]
hey, thanks for letting us know! will look into the issue and see what went wrong.
konart 1 days ago [-]
All TTS generations are too fast. It's almost I'm listening to a podcast on 1.25-1.5x speed.
toebee 1 days ago [-]
thanks for the feedback! will investigate and get it fixed
iharnoor 2 days ago [-]
By next month the competition for TTS will be even more!
Voice models are not winner take all market unlike LLM APIs
I meant the Nari inference engine for Qwen3-ASR. I'm aware that Qwen3-ASR is open source, but I don't see a repo under https://github.com/nari-labs for nari-qwen3-asr or similar.
the qwen3-asr inference repo is not OSSed as of now. we're planning to write a paper or tech report on it as it contains some general techniques for ASR inference.
kshmir 1 days ago [-]
how do I follow you? I have a small 5090 doing inference all the time and I barely use tts but a lot of asr, mostly whisper, I ported your tech report for tts and implemented some improvements on my whisper inference based on your tech report as well!
would love to talk sometime!
verdverm 2 days ago [-]
They have a number of demos and examples in their HF space
I saw a local-ai demo (something + gemma), where the person used ASR to get text and gemma to clean it up (like turning "question mark" into a literal "?", bullet points another one). The presenter also showed a gemma only option, that did both in one go, but had a higher WER on average, and even though the formatting statements were handled without a multi-stage pipeline, they preferred the multi-stage overall
yoloakki 2 days ago [-]
You definitely need independent evals by Datapoint AI or someone who can verify your claims about TTS quality
mowmiatlas 2 days ago [-]
Cool, I’ve released something to the same beat of the dr this weekend as well
The horizontal moving elements of examples become stuck and unable to be scolled once one of them is played. I'm using Vivaldi (chrome based) on Android
Added it to my blind TTS model comparison leaderboard. So far Darwin TTS is the open model leading the pack, ElevenLabs is at the lead.
https://huggingface.co/zeropointnine/Darwin-TTS-1.7B-Cross-Q...
Probably far fetched for now, but I think the next big evolution is building the pareto/much cheaper alternative to GPT-Live-1.
The STT/TTS market is quite saturated, while today, there's almost no cheap/open source alternative to GPT-Live-1.
Getting something conversationally better has been done, the tool calling will likely be worse though.
The infrastructure for real time is really annoying though.
+ Video is unlikely unless they are willing to give up margins.
Didn't take long at all for Google to smash out with 50% lol
For OP the clip name is nari-nina-01a0a12f-980a-765e-8029-fa56bd23210d.wav
Voice models are not winner take all market unlike LLM APIs
Coming here as Developer Relations at AssemblyAI
Is the ASR inference engine open source as well?
The Huggingface link on https://narilabs.com/product/stt/ links to https://huggingface.co/Qwen/Qwen3-ASR-1.7B , not anything under https://huggingface.co/nari-labs
would love to talk sometime!
https://huggingface.co/Qwen/spaces
I saw a local-ai demo (something + gemma), where the person used ASR to get text and gemma to clean it up (like turning "question mark" into a literal "?", bullet points another one). The presenter also showed a gemma only option, that did both in one go, but had a higher WER on average, and even though the formatting statements were handled without a multi-stage pipeline, they preferred the multi-stage overall
https://github.com/loudreader/loudkit
I think real time natural tts should be possible everywhere soon