TOtosh·Oct 11, 2023Submission
Silero Models: pre-trained speech-to-text, text-to-speech and text-enhancement
4 points · Open discussionsnakers4
silero-models
Jupyter Notebook2020–2023
snakers4 / silero-models
silero-models
Silero Models: pre-trained text-to-speech models made embarrassingly simple
12Distinct people
20Mentions
0Explicit recommenders
6.2KGitHub stars
What to know before exploring
What it is
Silero Models: pre-trained text-to-speech models made embarrassingly simple
GitHub
Where it is used
No documented product use was verified in the last source check.
What people are saying
No recommendation phrase was found in the indexed comments.
No classified warning is attached to this page. Read the discussions before treating silence as approval.
Also in these discussions
No other repository is repeatedly linked in the same discussions.
This week on Hacker News
Nov 7–Nov 14 UTC · Direct repository links and separately labeled related project stories. Indexed through 11/14/2026, 5:04:57 PM UTC.
No direct repository links were found in the indexed HN discussions over the past seven days. Older discussions are available below.
Mentions over time
HN mentions by year20 dated mentions
Observed mentions: 2020-09-18 07:50:24 UTC to 2023-10-11 15:20:01 UTC.
Counts come from the repository’s dated source evidence. Index cutoff: 2026-11-02 08:16:59 UTC.
View mention counts and date ranges
| Period | Start (UTC) | End (exclusive, UTC) | Mentions | Submissions | Comments |
|---|---|---|---|---|---|
| 2020 | 2020-01-01 00:00:00 UTC | 2021-01-01 00:00:00 UTC | 3 | 2 | 1 |
| 2021 | 2021-01-01 00:00:00 UTC | 2022-01-01 00:00:00 UTC | 5 | 0 | 5 |
| 2022 | 2022-01-01 00:00:00 UTC | 2023-01-01 00:00:00 UTC | 10 | 1 | 9 |
| 2023 | 2023-01-01 00:00:00 UTC | 2024-01-01 00:00:00 UTC | 2 | 1 | 1 |
What Hacker News says
20 sourcesPOST-EDIT, CORRECTED ANSWER
I doubt it's currently actually "the best open source text to speech", but the answer I came up with when throwing a couple of hours at the problem some months ago was "ttsprech" [3].
Following the guide, it was pretty trivial to make the model render my sample text in about 100 English "voices" (many of which were similar to each other, and in varying quality). Sampling those, I got about 10 that were pretty "good". And maybe 6 that were the "best ones" (very natural, not annoying to listen to, actually sounded like a person by and large), and maybe 2 made the top (as in, a tossup for the most listenable, all factors considered).
IIRC, the license was free for noncommercial use only. I'm not sure exactly "how open source" they are, but it was simple to install the dependencies and write the basic Python to
Read the rest
try it out; I had to write a for loop to try all the voices like I wanted. I ended using something else for the project for other reasons, but this could still be a fairly good backup option for some use cases, IMO.
PRE-EDIT, ERRONEOUS ANSWER
Same as above, but I had said "Silero" [0, 1, 2] originally, which I started trying out too, before switching to a third (less open) option.
[0] https://github.com/snakers4/silero-models#text-to-speech
[1] https://silero.ai
[2] https://github.com/snakers4/silero-models#standalone-use
[3] https://github.com/Grumbel/ttsprech#usage
My own experience agrees: the generally available "SOTA" models are not especially robust, and can be _extremely_ bad (>50% absolute error rate) at some tasks. I'll post some preliminary numbers in a sibling comment and look into running my full set of tests on Whisper.
It looks like Whisper is probably leaving a lot of accuracy on the table, but initially it does seem to be a lot more robust than general "SOTA" models.
For a quick comparison, Silero's accuracy charts are kind of nice because they post results for a large variety of datasets. Scroll down to the EN V6 xlarge EE model (not the xlarge CE) [1]
[1] https://github.com/snakers4/silero-models/wiki/Quality-Bench...
In Whisper – open source speech recognition by OpenAISome months ago I tried the Silero Models: https://github.com/snakers4/silero-models
With the audio sources I had, in English, the transcription had many mistakes. The good side is that installing and running the software worked as described in their documentation, so maybe it’s worth giving it a try by yourself.
In Ask HN: Open-source video transcribing software?What does CC BY-NC-SA actually mean when applied to the model? Are the .wav files generated by it considered CC BY-NC-SA as well or only software derived from it? Would I be allowed to use the output in say a Youtube video?
PS: This still points to the old AGPL license: https://github.com/snakers4/silero-models/wiki/Licensing-and...
In Silero V3: fast high-quality text-to-speech in 20 languages with 173 voicesBased on my limited understanding it seems to be possible to use (Not training) these models using C# and the ONNX runtime without needing any Python or other dependencies installed (would need to download the models manually). https://github.com/snakers4/silero-models#onnx=
Is this correct?
In Silero V3: fast high-quality text-to-speech in 20 languages with 173 voicesHN removes the HTML anchors.
The link (which I also copied below in the comments, b/c it was noticed after publishing) should be:
- https://github.com/snakers4/silero-models#text-to-speech
This link leads to the TTS section, whish contains the full list of all of the speakers and languages.
In Silero V3: fast high-quality text-to-speech in 20 languages with 173 voiceshttps://github.com/snakers4/silero-models/tree/941f911858f51...
Looks like it was previously GPL until 2 months ago, you could use the old version
In Silero V3: fast high-quality text-to-speech in 20 languages with 173 voicesCan you elaborate further? I am not familiar with the field, but their benchmarks here seem to show quality similar to Google: https://github.com/snakers4/silero-models/wiki/Quality-Bench...
The only trick I can see being played is that Google was benchmarked on September 2020, so likely has already improved and they don't want to show that. Is CommonVoice a better standard to use when comparing these tools?
In Silero V3: fast high-quality text-to-speech in 20 languages with 173 voicesAlso also, HN cuts the anchors in ULRs, so here is the full URL
- https://github.com/snakers4/silero-models#text-to-speech
In Silero V3: fast high-quality text-to-speech in 20 languages with 173 voicesSilero V3: fast high-quality text-to-speech in 20 languages with 173 voices
256 points · Open discussionSilero[0] seems to have decent performance (although you will have to some minimal coding). I believe there are better ones if you're willing to tinker a bit more.
[0]: https://github.com/snakers4/silero-models
In Otter.ai has saved reporters hours transcribing interviews. Caveat emptorFast STT models rivaling Google
- https://github.com/snakers4/silero-models#speech-to-text
- https://github.com/snakers4/silero-models/wiki/Quality-Bench...
- https://habr.com/ru/post/519562/
Disclaimer, this is my project
In [dead]Check out silero-models, available in colab with 10 speakers, running 3-4 RTS on one CPU core:
- https://github.com/snakers4/silero-models#text-to-speech
- https://habr.com/ru/post/549482/
In [dead]You should also totally check out silero-models, which are also available in colab with 10 speakers:
- https://github.com/snakers4/silero-models#text-to-speech - https://habr.com/ru/post/549482/
Disclaimer, this is my independent project
In TTS: Text-to-Speech for AllShowing the 15 most recent of 20 sources. Enable JavaScript to search and browse every indexed mention.