(Verified) suggestions for more sites

[tobybear]

This is not an actual "issue", just a list of websites offering TTS voices that I have already successfully implemented for rendering TTS audio using simple internet requests (similar to the ones already available in this project). No accounts or access keys are needed for these, and the list has now grown to 100 engines/sites (all verified to be working as of today), offering more than 24.000 voices.
This is just in case you are looking for inspiration for more sites/voices to add (and with the certainty that these can actually be accessed using normal requests).

[ 0]: 'GoogleTranslate' [64 voices, 2 parameters] [http://translate.google.com/]
[ 1]: 'Festival' [18 voices, 2 parameters, Festival] [https://www.cstr.ed.ac.uk/projects/festival/]
[ 2]: 'Cerence' [138 voices, 1 parameter, Nuance Vocalizer] [https://codefactoryglobal.com/]
[ 3]: 'Sestek' [24 voices, 2 parameters] [https://ttsdemo.sestek.com/]
[ 4]: 'SpeechGen' [694 voices, 2 parameters, Amazon Polly, Microsoft Azure] [https://speechgen.io]
[ 5]: 'iSpeech' [40 voices, 2 parameters] [http://www.ispeech.org/]
[ 6]: 'VoiceRSS API' [49 voices, 4 parameters] [http://www.voicerss.org/]
[ 7]: 'VoiceRSS WEB' [49 voices, 2 parameters] [https://voicerss.org/]
[ 8]: 'Responsive Voice' [98 voices, 5 parameters] [https://responsivevoice.org/]
[ 9]: 'TextToMP3' [25 voices, 2 parameters] [https://www.texttomp3.online/]
[ 10]: 'Text2Speech' [6 voices, 3 parameters] [https://www.text2speech.org/]
[ 11]: 'IBM Watson' [30 voices, 4 parameters, IBM Watson] [https://www.ibm.com/demos/live/tts-demo/]
[ 12]: 'TTSMP3' [60 voices, 2 parameters, Amazon Polly] [https://www.ttsmp3.com/]
[ 13]: 'InfoRobo' [60 voices, 4 parameters] [https://inforobo.com/]
[ 14]: 'PlayHT' [829 voices, 5 parameters, Amazon Polly, Google Wavenet, IBM Watson, Microsoft Azure] [https://play.ht/]
[ 15]: 'TTSRobotRU' [46 voices, 4 parameters, Yandex] [http://texttospeechrobot.com/]
[ 16]: 'VoxWorker' [2 voices, 4 parameters] [https://voxworker.com/]
[ 17]: 'Acapela' [239 voices, 2 parameters] [https://www.acapela-group.com/]
[ 18]: 'AcapelaDemo' [1 voice, 2 parameters] [https://www.acapela-group.com/]
[ 19]: 'Cepstral' [27 voices, 5 parameters] [https://www.cepstral.com/]
[ 20]: 'ODO' [220 voices, 5 parameters, Espeak, MBROLA] [https://odo.lv/]
[ 21]: 'Cereproc' [134 voices, 2 parameters] [https://www.cereproc.com/]
[ 22]: 'TTSTool' [120 voices, 2 parameters, Amazon Polly, Microsoft Azure] [https://ttstool.com/]
[ 23]: 'TTSOnline' [12 voices, 4 parameters] [https://texttospeech.onl/]
[ 24]: 'Tetyys' [29 voices, 4 parameters, TruVoice, SAPI] [https://tetyys.com/]
[ 25]: 'JTalk' [6 voices, 5 parameters] [http://open-jtalk.sp.nitech.ac.jp/]
[ 26]: 'FliteHTS' [1 voice, 5 parameters, Festival] [http://flite-hts-engine.sp.nitech.ac.jp/]
[ 27]: 'ImTranslator' [11 voices, 3 parameters] [https://text-to-speech.imtranslator.net/]
[ 28]: 'Linguatec' [145 voices, 4 parameters, Nuance Vocalizer] [https://www.linguatec.de/]
[ 29]: 'Alfanum' [25 voices, 4 parameters] [https://www.alfanum.co.rs/]
[ 30]: 'SpeechTech' [15 voices, 2 parameters] [https://www.speechtech.cz/]
[ 31]: 'MelroseLabs' [60 voices, 2 parameters, Amazon Polly] [https://melroselabs.com/]
[ 32]: 'Listnr' [791 voices, 4 parameters, Amazon Polly, Google Wavenet, Microsoft Azure] [https://www.listnr.tech/]
[ 33]: 'ElevenLabs' [44 voices, 2 parameters] [https://beta.elevenlabs.io/]
[ 34]: 'Bhashini' [31 voices, 2 parameters] [https://tts.bhashini.ai/]
[ 35]: 'Oddcast' [288 voices, 3 parameters] [https://www.oddcast.com/]
[ 36]: 'Overdub' [8 voices, 2 parameters] [https://public.lyrebird.ai/]
[ 37]: 'Fasthub' [66 voices, 6 parameters] [https://fasthub.net/]
[ 38]: 'MicMonster' [488 voices, 2 parameters, Microsoft Azure] [https://micmonster.com/]
[ 39]: 'Uberduck' [41 voices, 2 parameters] [https://uberduck.ai/]
[ 40]: 'FakeYou' [3985 voices, 2 parameters] [https://fakeyou.com/]
[ 41]: 'NoteVibes' [181 voices, 2 parameters, Google Wavenet] [https://notevibes.com/]
[ 42]: 'RHVoice' [94 voices, 6 parameters] [https://data2data.ru/tts/]
[ 43]: 'VoiceMod' [21 voices, 3 parameters] [https://tuna.voicemod.net/text-to-speech]
[ 44]: 'BingTranslator' [313 voices, 4 parameters, Microsoft Azure] [https://www.bing.com/translator]
[ 45]: 'VoiceForge' [26 voices, 2 parameters] [https://voiceforge.com/]
[ 46]: 'Streamlabs' [60 voices, 2 parameters] [https://streamlabs.com/]
[ 47]: 'StreamElements' [205 voices, 2 parameters] [https://streamelements.com/]
[ 48]: 'OpenAI' [6 voices, 2 parameters] [https://platform.openai.com/]
[ 49]: 'ModelTalker' [20 voices, 2 parameters] [https://www.modeltalker.org/]
[ 50]: 'ReadTheWords' [15 voices, 2 parameters] [https://www.readthewords.com/]
[ 51]: 'ReadSpeaker' [132 voices, 2 parameters] [https://www.readspeaker.com/]
[ 52]: 'TTSFree' [731 voices, 2 parameters, Google Wavenet, Microsoft Azure] [https://ttsfree.com/]
[ 53]: 'Baidu' [6 voices, 4 parameters] [https://ai.baidu.com/tech/speech/tts_online]
[ 54]: 'VolcTrans' [36 voices, 2 parameters] [https://translate.volcengine.com/]
[ 55]: 'iMyPhone' [2564 voices, 2 parameters] [https://www.imyfone.com/]
[ 56]: 'GLaDOS' [1 voice, 2 parameters] [https://glados.c-net.org/]
[ 57]: 'AudioBot' [484 voices, 2 parameters, Google Wavenet, Microsoft Azure] [https://audio-bot.com/]
[ 58]: 'MP3TTS' [1255 voices, 7 parameters] [https://mp3tts.com/]
[ 59]: 'TextToVoice' [791 voices, 3 parameters, Amazon Polly, Google Wavenet, Microsoft Azure] [https://www.texttovoice.online/]
[ 60]: 'AnyToSpeech' [257 voices, 2 parameters, Google Wavenet] [https://anytospeech.com/]
[ 61]: 'Youdao' [58 voices, 3 parameters] [https://youdao.com/]
[ 62]: 'YoudaoDict' [37 voices, 2 parameters] [http://dict.youdao.com/]
[ 63]: 'SpanishDict' [2 voices, 2 parameters] [https://spanishdict.com/]
[ 64]: 'Yandex' [2 voices, 2 parameters, Yandex] [http://tts.voicetech.yandex.net/tts]
[ 65]: 'Naver' [16 voices, 5 parameters] [https://naver.com/]
[ 66]: 'Speechify' [221 voices, 2 parameters, Google Wavenet, Microsoft Azure, Speechify] [https://speechify.com/]
[ 67]: 'MWSReader' [26 voices, 2 parameters] [https://www.mwsreader.com/]
[ 68]: 'NextUp' [45 voices, 2 parameters, Ivona] [https://nextup.com/]
[ 69]: 'TTSMagic' [150 voices, 2 parameters, Amazon Polly, Google Wavenet, Microsoft Azure] [https://www.ttsmagic.com/]
[ 70]: 'Gesserit' [112 voices, 2 parameters, TikTok] [https://gesserit.co/]
[ 71]: 'Speech2Go' [47 voices, 2 parameters, Amazon Polly] [https://speech2go.online/]
[ 72]: 'TikTok' [112 voices, 2 parameters, TikTok] [https://weilbyte.github.io/tiktok-tts/]
[ 73]: 'Vidnoz' [313 voices, 2 parameters, Microsoft Azure] [https://www.vidnoz.com/]
[ 74]: 'Veed.io' [565 voices, 2 parameters, Microsoft Azure] [https://www.veed.io/]
[ 75]: 'Voicebooking' [428 voices, 2 parameters, Google Wavenet] [https://www.voicebooking.com/]
[ 76]: '101 Soundboards' [4720 voices, 2 parameters] [https://www.101soundboards.com/]
[ 77]: 'Crikk' [313 voices, 2 parameters, Microsoft Azure] [https://crikk.com/text-to-speech/]
[ 78]: 'Aquest' [36 voices, 3 parameters] [https://www.a-quest.com/]
[ 79]: 'Cyzon' [1 voice, 2 parameters, DECTalk] [https://tts.cyzon.us/]
[ 80]: 'SoundOfText' [61 voices, 2 parameters, Google Wavenet] [https://soundoftext.com/]
[ 81]: 'FluxToucan' [12 voices, 4 parameters] [https://flux9665-ims-toucan.hf.space/]
[ 82]: 'WhisperSpeech' [2 voices, 2 parameters] [https://replicate.com/lucataco/whisperspeech-small]
[ 83]: 'HierSpeechPP' [1 voice, 2 parameters] [https://replicate.com/adirik/hierspeechpp]
[ 84]: 'TortoiseTTS' [32 voices, 3 parameters] [https://replicate.com/afiaka87/tortoise-tts]
[ 85]: 'CoquiXTTS' [1 voice, 3 parameters] [https://replicate.com/lucataco/xtts-v2]
[ 86]: 'StyleTTS2' [1 voice, 2 parameters] [https://replicate.com/adirik/styletts2]
[ 87]: 'Talkify' [246 voices, 5 parameters] [https://talkify.net/]
[ 88]: 'Bark' [1 voice, 2 parameters] [https://replicate.com/suno-ai/bark]
[ 89]: 'Google Pico' [6 voices, 4 parameters] [https://android.googlesource.com/]
[ 90]: 'C64 SAM' [1 voice, 3 parameters] [https://discordier.github.io/sam/]
[ 91]: 'Ronan' [5 voices, 6 parameters] [https://www.pouet.net/prod.php?which=15073]
[ 92]: 'Piper' [3 voices, 2 parameters] [https://github.com/rhasspy/piper]
[ 93]: 'Sherpa' [109 voices, 2 parameters] [https://github.com/k2-fsa/sherpa-onnx]
[ 94]: 'EspeakNG' [93 voices, 4 parameters] [https://github.com/espeak-ng/]
[ 95]: 'STSpeech' [2 voices, 3 parameters] [https://www.atariuptodate.de/en/1686/st-speech]
[ 96]: 'GnuSpeech' [5 voices, 4 parameters] [https://en.wikipedia.org/wiki/Gnuspeech]
[ 97]: 'DECTalk' [9 voices, 3 parameters, DECTalk] [https://en.wikipedia.org/wiki/DECtalk]
[ 98]: 'RSynth' [1 voice, 4 parameters] [https://rsynth.sourceforge.net/]
[ 99]: 'Balcon' [14 voices, 4 parameters, SAPI] [https://www.cross-plus-a.com/balabolka.htm]
--> 100 engines, 24026 voices

[HappyMac3920]

There are quite a few issues with some of these TTS demo websites:

First, there are some websites that are rate limited (like only 500 characters per IP address per day), like TTSMP3 and Cepstral.

Second, there are websites that require an API key and credits to generate speech, like OpenAI.

Third, Cereproc voices used to be included on this website, however Cereproc contacted Chris to remove the voices from his website.

[tobybear]

Of course there are usage restrictions for pretty all of these sites to prevent abuse, which is totally understandable. The same is true however for the current sites being used, as these are all provided for evaluation purposes only.
TTSMP3 has 3000 characters per day, for Cepstral I didn't run into a limit yet but didn't try hard tbh.
All sites listes however can be used without requiring a personalized API key, including OpenAI (which has an inofficial generic API key). But the more people are using these sites for free, the less likely they will be available in future, or will require personal/restricted/paid accounts for sure.
I know about Cereproc contacting Chris and that they do now want to appear here, which I totally respect.

[HappyMac3920]

Of course there are usage restrictions for pretty all of these sites to prevent abuse, which is totally understandable. The same is true however for the current sites being used, as these are all provided for evaluation purposes only.

TTSMP3 has 3000 characters per day, for Cepstral I didn't run into a limit yet but didn't try hard tbh.

All sites listes however can be used without requiring a personalized API key, including OpenAI (which has an inofficial generic API key). But the more people are using these sites for free, the less likely they will be available in future, or will require personal/restricted/paid accounts for sure.

I know about Cereproc contacting Chris and that they do now want to appear here, which I totally respect.

For OpenAI, what is the API key?
Or is it private?

[tobybear]

It is sort of publicly private :)
There are several OpenAI demo sites that use a static authentication bearer token which can be used from anywhere.
Also, last time I looked, several model/spaces using OpenAI TTS on huggingface.co had tokens/keys filled in, although restrictions for using these also apply here.